INDEX

Explanations

numeric information such as ratings, scores, or quantities

oai_token-act-pair · gpt-3.5-turbo

phrases related to comparison and evaluation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

©¶æ¥µ

-0.82

agram

-0.81

DM

-0.74

paper

-0.72

Catal

-0.69

outside

-0.69

je

-0.68

aga

-0.66

SQ

-0.66

 Saga

-0.65

POSITIVE LOGITS

 positives

1.15

 thrive

1.11

 successes

1.07

 positive

0.96

 recover

0.95

 hopeful

0.94

 heroism

0.93

 constructive

0.93

 triumphant

0.92

 liberating

0.92

Activations Density 0.382%

numeric information such as ratings, scores, or quantities

phrases related to comparison and evaluation

No Comments

No Known Activations