INDEX

Explanations

names and terms related to individuals or places

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Fran" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

accompanied

-0.82

gow

-0.78

eenth

-0.77

 tomat

-0.75

wrapper

-0.75

scape

-0.74

een

-0.68

Tokens

-0.66

footed

-0.65

 stakes

-0.65

POSITIVE LOGITS

 Dres

1.12

 Fran

1.09

kel

0.94

furt

0.93

zen

0.91

Fran

0.90

Vog

0.84

opol

0.82

ois

0.80

Ã§

0.80

Activations Density 0.010%

names and terms related to individuals or places

mentions of the name "Fran" and its variations

No Comments

No Known Activations