INDEX

Explanations

statements or scenarios related to potential consequences or outcomes

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "would be."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ciating

-0.71

plex

-0.62

enary

-0.59

iop

-0.59

 practise

-0.58

 process

-0.58

anwhile

-0.57

 dispute

-0.57

onential

-0.56

 promise

-0.56

POSITIVE LOGITS

fall

0.97

 able

0.94

 considered

0.87

acons

0.85

fitting

0.83

falls

0.82

hemoth

0.81

heading

0.80

 taken

0.78

 judged

0.77

Activations Density 0.159%

statements or scenarios related to potential consequences or outcomes

instances of the phrase "would be."

No Comments

No Known Activations