INDEX

Explanations

words related to stress and its effects

oai_token-act-pair · gpt-3.5-turbo

references to stress

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

izons

-0.77

ettes

-0.69

pires

-0.65

kie

-0.65

advertising

-0.64

Aust

-0.63

HCR

-0.62

 merger

-0.61

gewater

-0.61

*/(

-0.60

POSITIVE LOGITS

stress

1.02

 stress

1.00

 Stress

0.94

ors

0.88

 reliever

0.85

âĶģ

0.80

 cooker

0.79

urer

0.74

otics

0.71

uler

0.70

Activations Density 0.006%

words related to stress and its effects

references to stress

No Comments

No Known Activations