INDEX

Explanations

strong emotional reactions or extreme evaluations

oai_token-act-pair · gpt-3.5-turbo

expressions of strong negative opinions or experiences

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Fukushima

-0.73

eric

-0.71

 Daylight

-0.65

 Indust

-0.64

iencies

-0.62

ews

-0.61

 Kimber

-0.59

ATF

-0.59

Tsu

-0.59

itu

-0.58

POSITIVE LOGITS

 EVER

0.96

 ever

0.89

 feas

0.75

illion

0.74

 imaginable

0.73

Merit

0.73

undai

0.70

aid

0.69

Ever

0.66

!'

0.65

Activations Density 0.145%

strong emotional reactions or extreme evaluations

expressions of strong negative opinions or experiences

No Comments

No Known Activations