INDEX

Explanations

phrases related to recognizable names or entities

oai_token-act-pair · gpt-3.5-turbo

words related to language or communication

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

 Pilgrim

-0.73

 showc

-0.65

 Sloan

-0.63

 McDonnell

-0.60

 ende

-0.59

 McGee

-0.59

 Booker

-0.59

err

-0.57

 infertility

-0.57

days

-0.56

POSITIVE LOGITS

pta

0.99

arette

0.95

cci

0.90

vernment

0.88

illac

0.87

eteenth

0.86

arin

0.85

hou

0.82

ogo

0.81

arant

0.80

Activations Density 0.013%

phrases related to recognizable names or entities

words related to language or communication

No Comments

No Known Activations