INDEX

Explanations

mentions of government officials and actions

oai_token-act-pair · gpt-3.5-turbo

references to political figures and governmental roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

EStreamFrame

-0.70

$.

-0.65

}.

-0.57

ĸļ

-0.56

 respectively

-0.55

)).

-0.55

selves

-0.54

'.

-0.53

 breeze

-0.52

 checkout

-0.52

POSITIVE LOGITS

 spokesman

0.72

 acknowledges

0.63

 acknowledged

0.58

 reacted

0.58

 spokesperson

0.57

Ban

0.56

 apologised

0.55

 spokeswoman

0.55

Len

0.54

 declares

0.53

Activations Density 0.762%

mentions of government officials and actions

references to political figures and governmental roles

No Comments

No Known Activations