INDEX

Explanations

phrases discussing actions, decisions, or events with significant impact or consequences

oai_token-act-pair · gpt-3.5-turbo

affirmative statements about actions or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

Gur

-0.70

Naj

-0.69

Cyr

-0.65

 Nept

-0.65

 Cher

-0.64

 Pats

-0.62

Sly

-0.62

 Swed

-0.61

 Mush

-0.60

 Heal

-0.60

POSITIVE LOGITS

ushima

0.86

senal

0.83

 nevertheless

0.81

 nonetheless

0.76

ategory

0.75

 akin

0.73

SourceFile

0.71

ADRA

0.71

¿½

0.70

ŃĶ

0.70

Activations Density 0.669%

phrases discussing actions, decisions, or events with significant impact or consequences

affirmative statements about actions or conditions

No Comments

No Known Activations