INDEX

Explanations

phrases indicating actions that have happened or will happen

oai_token-act-pair · gpt-3.5-turbo

phrases indicating repeated actions or occurrences

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

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

Published

-0.67

arov

-0.65

cipled

-0.63

CHAT

-0.61

usterity

-0.59

eware

-0.57

asta

-0.56

isphere

-0.54

enh

-0.54

issued

-0.53

POSITIVE LOGITS

 elsewhere

0.94

 during

0.81

 when

0.79

 whenever

0.78

 throughout

0.78

pez

0.78

 ours

0.77

 before

0.76

 today

0.76

 with

0.71

Activations Density 0.078%

phrases indicating actions that have happened or will happen

phrases indicating repeated actions or occurrences

No Comments

No Known Activations