INDEX

Explanations

phrases related to labeling and classification

oai_token-act-pair · gpt-3.5-turbo

references to labels and labeling practices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

tein

-0.70

issance

-0.67

 Globe

-0.66

Yin

-0.64

vati

-0.64

ashington

-0.64

ctica

-0.63

gm

-0.63

olars

-0.62

hire

-0.62

POSITIVE LOGITS

 label

0.90

mates

0.89

 labels

0.87

mate

0.85

cloth

0.80

strip

0.76

label

0.76

Label

0.75

red

0.75

mark

0.74

Activations Density 0.022%

phrases related to labeling and classification

references to labels and labeling practices

No Comments

No Known Activations