INDEX

Explanations

well-known titles from media outlets or organizations

oai_token-act-pair · gpt-3.5-turbo

names of media outlets, organizations, and significant locations

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

ËĪ

-0.60

-0.57

="#

-0.54

norm

-0.52

jl

-0.51

ãĥ¼ãĥĨãĤ£

-0.51

 Bellev

-0.51

tight

-0.50

endum

-0.49

NON

-0.49

POSITIVE LOGITS

 respectively

1.60

 alike

1.31

Interstitial

0.79

etc

0.79

 versa

0.78

SPONSORED

0.78

0.69

 depending

0.69

 Aven

0.67

depending

0.63

Activations Density 0.559%

well-known titles from media outlets or organizations

names of media outlets, organizations, and significant locations

No Comments

No Known Activations