INDEX

Explanations

phrases indicating a transition or shift in focus

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

eton

-0.69

lde

-0.61

ifax

-0.60

en

-0.56

weet

-0.54

ania

-0.52

ource

-0.52

lie

-0.52

ens

-0.52

uran

-0.51

POSITIVE LOGITS

agy

0.56

 Gupta

0.56

chest

0.52

zu

0.49

ggle

0.49

usions

0.49

kered

0.47

 Sharma

0.47

Reviewer

0.46

 BALL

0.46

Activations Density 0.187%

phrases indicating a transition or shift in focus

No Comments

No Known Activations