INDEX

Explanations

phrases or words related to academic research studies

oai_token-act-pair · gpt-3.5-turbo

references to various studies and research

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

gettable

-0.99

interstitial

-0.85

 notor

-0.76

 sequest

-0.75

guyen

-0.74

tesy

-0.73

 lethal

-0.73

 landsl

-0.72

 warmer

-0.72

comed

-0.72

POSITIVE LOGITS

 Study

1.29

 Guides

1.10

 Interpret

1.02

 Group

1.00

 Concepts

0.97

 Plans

0.97

 Groups

0.96

Study

0.96

 Areas

0.94

 Scores

0.94

Activations Density 0.008%

phrases or words related to academic research studies

references to various studies and research

No Comments

No Known Activations