INDEX

Explanations

keywords related to numerical values

oai_token-act-pair · gpt-3.5-turbo

proper nouns or names and specific identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

bara

-0.76

letter

-0.75

din

-0.75

gal

-0.74

rament

-0.72

 Dominion

-0.67

 Letter

-0.67

 BALL

-0.66

 ANGEL

-0.66

ãĤ¼

-0.66

POSITIVE LOGITS

icho

1.03

ich

1.02

ico

0.92

icks

0.83

ickers

0.83

ij

0.82

ikan

0.80

ike

0.80

ick

0.78

ican

0.78

Activations Density 0.381%

keywords related to numerical values

proper nouns or names and specific identifiers

No Comments

No Known Activations