INDEX

Explanations

proper nouns followed by what seems to be a unique identifier

oai_token-act-pair · gpt-3.5-turbo

the letter 'P' in varying contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 hitch

-0.62

ties

-0.61

uality

-0.60

 galaxies

-0.59

 dressing

-0.58

 straw

-0.56

 editions

-0.54

 laps

-0.54

 unde

-0.54

 seams

-0.53

POSITIVE LOGITS

3.30

Ps

1.95

1.72

PF

1.71

PB

1.70

PP

1.62

PK

1.60

PT

1.47

PN

1.45

Activations Density 0.025%

proper nouns followed by what seems to be a unique identifier

the letter 'P' in varying contexts

No Comments

No Known Activations