INDEX

Explanations

mentions of individuals by their gender-neutral pronouns

oai_token-act-pair · gpt-3.5-turbo

references to individuals and their roles or achievements

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

Untitled

-0.73

 bothering

-0.71

Battery

-0.63

Enlarge

-0.62

mma

-0.61

âĺħâĺħ

-0.60

âĢ¢âĢ¢

-0.60

 pregn

-0.60

Leaks

-0.60

 clichÃ©

-0.60

POSITIVE LOGITS

 also

0.94

pherd

0.91

theless

0.88

miah

0.88

 consists

0.86

 consisted

0.84

 comprises

0.83

 graduated

0.81

ffield

0.80

'll

0.79

Activations Density 0.305%

mentions of individuals by their gender-neutral pronouns

references to individuals and their roles or achievements

No Comments

No Known Activations