INDEX

Explanations

attributions of statements to individuals

oai_token-act-pair · gpt-3.5-turbo

statements or assertions attributed to individuals followed by the phrase "he said" or "she said."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

wcsstore

-0.56

commit

-0.54

 litter

-0.53

protect

-0.51

wear

-0.51

fet

-0.51

bol

-0.51

 monog

-0.50

 coffin

-0.50

NetMessage

-0.50

POSITIVE LOGITS

 paraph

0.76

 explained

0.74

said

0.71

 said

0.70

 elaborated

0.68

 Said

0.66

 quoting

0.63

Said

0.63

 joked

0.62

 told

0.62

Activations Density 0.309%

attributions of statements to individuals

statements or assertions attributed to individuals followed by the phrase "he said" or "she said."

No Comments

No Known Activations