INDEX

Explanations

phrases related to specific activities or practices

oai_token-act-pair · gpt-3.5-turbo

instances of the word "in."

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĤ¼ãĤ¦ãĤ¹

-0.73

syn

-0.69

fram

-0.63

 Schwar

-0.63

iev

-0.63

steen

-0.62

 rarity

-0.60

EStreamFrame

-0.59

 palette

-0.59

âī

-0.58

POSITIVE LOGITS

nostic

0.80

aming

0.74

aned

0.70

antle

0.67

antha

0.66

ering

0.65

onite

0.65

bes

0.65

okÃ©

0.64

amer

0.64

Activations Density 0.000%

phrases related to specific activities or practices

instances of the word "in."

No Comments

No Known Activations