INDEX

Explanations

References to dominance or dominant entities

oai_token-act-pair · gpt-3.5-turbo

references to dominant entities or influences in various contexts

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

ridges

-0.84

raising

-0.80

hire

-0.77

chery

-0.75

bsite

-0.73

ruary

-0.72

govtrack

-0.72

deals

-0.71

endment

-0.70

atron

-0.69

POSITIVE LOGITS

 bidder

0.87

 force

0.81

 themes

0.81

 theme

0.79

 personalities

0.75

 personality

0.74

XI

0.72

 forces

0.72

 narratives

0.71

 figures

0.68

Activations Density 0.044%

References to dominance or dominant entities

references to dominant entities or influences in various contexts

No Comments

No Known Activations