INDEX

Explanations

names specifically associated with clothing or attire being removed

oai_token-act-pair · gpt-3.5-turbo

terms related to hairdressing or hairstyles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ãĥ£

-0.70

©¶æ

-0.65

gum

-0.63

 hemisphere

-0.62

lder

-0.62

 Witnesses

-0.60

 thirds

-0.59

 elig

-0.59

STON

-0.59

 doubling

-0.58

POSITIVE LOGITS

ions

1.15

ively

1.10

ional

1.04

IVE

0.99

entially

0.91

encer

0.90

entials

0.89

furt

0.88

itect

0.87

mann

0.85

Activations Density 0.014%

names specifically associated with clothing or attire being removed

terms related to hairdressing or hairstyles

No Comments

No Known Activations