INDEX

Explanations

information related to decisions or actions being taken

oai_token-act-pair · gpt-3.5-turbo

instances of the word "move" in various contexts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 sqor

-0.79

omial

-0.72

vae

-0.71

 attm

-0.64

 Koran

-0.63

errors

-0.63

ordon

-0.63

inges

-0.62

Condition

-0.61

IZE

-0.60

POSITIVE LOGITS

 toward

0.96

able

0.95

 towards

0.95

backs

0.94

ments

0.92

ment

0.81

over

0.80

rers

0.78

abouts

0.77

 forward

0.73

Activations Density 0.038%

information related to decisions or actions being taken

instances of the word "move" in various contexts

No Comments

No Known Activations