INDEX

Explanations

phrases related to guidelines, instructions, or rules

oai_token-act-pair · gpt-3.5-turbo

numeric values and their associated concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

xual

-0.75

ennes

-0.72

 Leilan

-0.68

,...

-0.64

 =================================================================

-0.64

utical

-0.64

 parting

-0.62

 concluding

-0.61

tsky

-0.60

ĸļ

-0.60

POSITIVE LOGITS

ÂłÂłÂłÂł

0.85

ÂłÂłÂłÂłÂłÂłÂłÂł

0.84

ÂłÂłÂł

0.83

Ì¶

0.76

ÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂł

0.71

Avg

0.69

oss

0.67

Database

0.66

Limit

0.64

Ability

0.63

Activations Density 0.203%

phrases related to guidelines, instructions, or rules

numeric values and their associated concepts

No Comments

No Known Activations