INDEX

Explanations

mentions of specific locations or events

oai_token-act-pair · gpt-3.5-turbo

prepositions and their usage in context

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

lez

-0.77

apeake

-0.74

upt

-0.69

etsk

-0.68

aughs

-0.66

 Offline

-0.66

anu

-0.64

.,"

-0.63

initely

-0.63

Russ

-0.62

POSITIVE LOGITS

 Genie

0.71

 reproduction

0.69

 Judge

0.65

 sorcery

0.63

DeL

0.61

 supervision

0.60

 Dominican

0.60

ATIVE

0.60

 unaccompanied

0.60

 Alter

0.59

Activations Density 0.875%

mentions of specific locations or events

prepositions and their usage in context

No Comments

No Known Activations