INDEX

Explanations

the end of a statement or text

oai_token-act-pair · gpt-3.5-turbo

punctuations and query-like sentences indicating uncertainty or transition

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 imperson

-0.82

oun

-0.78

oud

-0.78

 nickname

-0.72

pecially

-0.68

ouf

-0.68

 fitted

-0.68

 exempt

-0.68

unts

-0.67

 skew

-0.67

POSITIVE LOGITS

 Releases

0.91

 Despair

0.89

 Hopefully

0.87

 Stay

0.86

 Tomorrow

0.82

 Until

0.82

 Goodbye

0.81

 Prepare

0.80

 Promise

0.78

 Newsp

0.78

Activations Density 0.592%

the end of a statement or text

punctuations and query-like sentences indicating uncertainty or transition

No Comments

No Known Activations