INDEX

Explanations

mentions of hair-related descriptions or actions

oai_token-act-pair · gpt-3.5-turbo

references to hair

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

igmat

-0.70

isconsin

-0.69

 Meier

-0.65

ellig

-0.64

Critical

-0.64

ENTION

-0.63

éĹĺ

-0.62

 Gutenberg

-0.62

igm

-0.62

 Reloaded

-0.62

POSITIVE LOGITS

pins

1.16

 hair

1.11

bread

1.10

 foll

1.01

pin

1.01

beard

0.99

brush

0.99

dress

0.97

bone

0.91

pieces

0.91

Activations Density 0.025%

mentions of hair-related descriptions or actions

references to hair

No Comments

No Known Activations