INDEX

Explanations

references to holding law enforcement accountable for their actions

oai_token-act-pair · gpt-3.5-turbo

instances of special characters or symbols that might indicate emphasis or sentiment

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

 Negro

-0.62

 surv

-0.60

gib

-0.58

 retreat

-0.56

 litter

-0.55

 catches

-0.55

jog

-0.55

 overlook

-0.55

ces

-0.54

 Roma

-0.54

POSITIVE LOGITS

ï¸ı

1.16

_>

0.84

0.83

SHIP

0.83

ï¸

0.81

sbm

0.81

except

0.79

selves

0.75

iversary

0.75

sure

0.75

Activations Density 0.365%

references to holding law enforcement accountable for their actions

instances of special characters or symbols that might indicate emphasis or sentiment

No Comments

No Known Activations