INDEX

Explanations

references to people or entities and actions related to their roles or influence

oai_token-act-pair · gpt-3.5-turbo

references to personal opinions or evaluations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Newsletter

-0.68

ocular

-0.61

ospital

-0.61

Doctors

-0.61

 chlorine

-0.60

 haircut

-0.59

 telephone

-0.58

 hospital

-0.58

cut

-0.57

equ

-0.57

POSITIVE LOGITS

etheless

1.06

genre

0.75

 indie

0.72

 fandom

0.72

 genuinely

0.71

 millenn

0.70

 Capcom

0.70

 nostalgia

0.69

 hopefully

0.69

 Metroid

0.68

Activations Density 1.720%

references to people or entities and actions related to their roles or influence

references to personal opinions or evaluations

No Comments

No Known Activations