INDEX

Explanations

proper nouns related to various individuals named Parker

oai_token-act-pair · gpt-3.5-turbo

references to the name "Parker."

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

ľ

-0.87

ornia

-0.86

ugal

-0.83

Ĳ

-0.76

ħĭ

-0.76

natureconservancy

-0.76

orical

-0.75

lex

-0.75

icter

-0.73

urg

-0.72

POSITIVE LOGITS

 Parker

0.97

 Brothers

0.95

 Bros

0.94

ville

0.89

 Chun

0.88

 Alley

0.88

inson

0.85

 Pose

0.80

stown

0.80

Park

0.78

Activations Density 0.015%

proper nouns related to various individuals named Parker

references to the name "Parker."

No Comments

No Known Activations