INDEX

Explanations

phrases related to instructions or specifications

oai_token-act-pair · gpt-3.5-turbo

references to intentions or purpose

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

bara

-0.59

itas

-0.58

Calling

-0.54

EMA

-0.53

mie

-0.50

Ted

-0.49

docker

-0.49

Invest

-0.49

erity

-0.49

 Vince

-0.49

POSITIVE LOGITS

 themselves

1.00

 individually

0.93

 clustered

0.85

 collectively

0.78

selves

0.78

 interchangeable

0.73

 interchange

0.73

 selves

0.73

 numbered

0.72

geries

0.71

Activations Density 1.227%

phrases related to instructions or specifications

references to intentions or purpose

No Comments

No Known Activations