INDEX

Explanations

people and organizations mentioned in news or media articles

oai_token-act-pair · gpt-3.5-turbo

references to various media outlets and news organizations

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

 obser

-0.62

 subdiv

-0.57

pires

-0.57

 proven

-0.56

 SHALL

-0.56

 veter

-0.54

pired

-0.54

 deserted

-0.54

 productive

-0.54

viz

-0.53

POSITIVE LOGITS

 afterward

0.81

via

0.79

 interviewer

0.79

0.78

ono

0.74

."

0.72

!.

0.72

 shortly

0.71

 afterwards

0.70

 yesterday

0.69

Activations Density 0.088%

people and organizations mentioned in news or media articles

references to various media outlets and news organizations

No Comments

No Known Activations