INDEX

Explanations

adjectives describing the intensity or quality of something

oai_token-act-pair · gpt-3.5-turbo

phrases indicating low quality or lack of effectiveness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

YES

-0.74

ilts

-0.71

Leaks

-0.66

çļ

-0.65

pez

-0.65

probably

-0.63

 Potential

-0.61

Ĥİ

-0.60

origin

-0.59

kamp

-0.57

POSITIVE LOGITS

 anymore

1.14

 glamorous

0.94

 flashy

0.93

 forgiving

0.90

 bothered

0.86

 bother

0.84

 conducive

0.81

nor

0.81

 appealing

0.80

 pleasant

0.79

Activations Density 0.064%

adjectives describing the intensity or quality of something

phrases indicating low quality or lack of effectiveness

No Comments

No Known Activations