INDEX

Explanations

phrases indicating exclusivity or specificity

oai_token-act-pair · gpt-3.5-turbo

phrases focused on exclusive or specialized categories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ealous

-0.82

REDACTED

-0.81

è¦ļéĨĴ

-0.80

KEN

-0.78

enhagen

-0.76

pid

-0.74

amen

-0.74

abul

-0.73

Ã©n

-0.73

someone

-0.73

POSITIVE LOGITS

 edition

0.98

 versions

0.90

 editions

0.88

 version

0.84

 equivalents

0.77

 television

0.77

 amph

0.74

 newspaper

0.74

ale

0.74

 clothing

0.73

Activations Density 0.204%

phrases indicating exclusivity or specificity

phrases focused on exclusive or specialized categories

No Comments

No Known Activations