INDEX

Explanations

mentions of clothing items, specifically shirts

oai_token-act-pair · gpt-3.5-turbo

references to T-shirts and clothing items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

FD

-0.65

 problems

-0.63

rel

-0.63

 inaccur

-0.63

 fixed

-0.62

inf

-0.62

 loopholes

-0.62

 cycles

-0.61

LD

-0.61

 established

-0.61

POSITIVE LOGITS

shirt

4.12

shirts

3.09

 Shirt

1.92

hirt

1.91

 shirt

1.70

coat

1.44

 shirts

1.42

dress

1.21

 jersey

1.15

sheet

1.11

Activations Density 0.014%

mentions of clothing items, specifically shirts

references to T-shirts and clothing items

No Comments

No Known Activations