INDEX

Explanations

mentions of significant actions or decisions being taken

oai_token-act-pair · gpt-3.5-turbo

references to significant actions or decisions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

omial

-0.73

IZE

-0.68

Condition

-0.67

Mub

-0.65

ussion

-0.58

icum

-0.58

 resid

-0.58

 havoc

-0.56

 Degree

-0.56

 Hath

-0.56

POSITIVE LOGITS

 toward

1.11

 towards

1.09

backs

0.95

able

0.94

abouts

0.87

 forward

0.87

overs

0.86

over

0.86

 away

0.85

offs

0.83

Activations Density 0.036%

mentions of significant actions or decisions being taken

references to significant actions or decisions

No Comments

No Known Activations