INDEX

Explanations

words related to a specific type of notation, particularly prefixes used in mathematical and computer sciences

oai_token-act-pair · gpt-3.5-turbo

terms related to hierarchical classifications or categorization systems, particularly prefixes

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

stood

-0.82

izen

-0.81

estern

-0.73

 Weir

-0.70

ARDS

-0.69

TIME

-0.69

å°Ĩ

-0.68

bold

-0.68

 Commons

-0.68

icago

-0.68

POSITIVE LOGITS

hesis

0.91

rontal

0.79

roleum

0.75

hetically

0.74

icip

0.74

hetical

0.73

requisite

0.71

 spoil

0.71

hend

0.68

heses

0.68

Activations Density 0.026%

words related to a specific type of notation, particularly prefixes used in mathematical and computer sciences

terms related to hierarchical classifications or categorization systems, particularly prefixes

No Comments

No Known Activations