INDEX

Explanations

events or achievements that are significant in history, such as being the first in a particular category or a major development in a field

oai_token-act-pair · gpt-3.5-turbo

terms related to first occurrences or notable distinctions in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tnc

-0.71

utics

-0.68

uts

-0.64

asty

-0.64

ĺħ

-0.63

 unfocusedRange

-0.62

iland

-0.62

cdn

-0.61

 Diseases

-0.61

beaut

-0.61

POSITIVE LOGITS

 imaginable

0.88

 foray

0.83

oslov

0.81

 outing

0.77

 ever

0.77

 EVER

0.75

achable

0.74

-|

0.71

 hurdle

0.70

istani

0.68

Activations Density 0.217%

events or achievements that are significant in history, such as being the first in a particular category or a major development in a field

terms related to first occurrences or notable distinctions in various contexts

No Comments

No Known Activations