INDEX

Explanations

references to hair-related actions and features

oai_token-act-pair · gpt-3.5-turbo

references to hair and its importance or characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ellig

-0.69

umbn

-0.67

eous

-0.65

VERTIS

-0.63

PsyNetMessage

-0.62

 Martial

-0.61

 Jude

-0.61

 Reloaded

-0.61

 Stafford

-0.60

sych

-0.60

POSITIVE LOGITS

 hair

1.15

bread

0.99

hair

0.98

 foll

0.97

pins

0.93

brush

0.93

 hairs

0.93

beard

0.92

 mustache

0.91

wig

0.90

Activations Density 0.013%

references to hair-related actions and features

references to hair and its importance or characteristics

No Comments

No Known Activations