INDEX

Explanations

specific years

oai_token-act-pair · gpt-3.5-turbo

references to the year 2005 or similar contextual markers associated with it

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

books

-0.87

ly

-0.84

cut

-0.80

iris

-0.80

eki

-0.80

icone

-0.79

fty

-0.77

osal

-0.75

swick

-0.75

ugu

-0.74

POSITIVE LOGITS

hips

0.82

marked

0.78

APH

0.75

 ILCS

0.69

ipolar

0.68

axies

0.65

ihar

0.64

 chrom

0.62

acy

0.61

COLOR

0.61

Activations Density 0.071%

specific years

references to the year 2005 or similar contextual markers associated with it

No Comments

No Known Activations