INDEX

Explanations

words related to different geographic regions, especially Wales

oai_token-act-pair · gpt-3.5-turbo

references to Wales and Welsh identity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ographed

-0.89

nels

-0.82

ebin

-0.74

oses

-0.71

analy

-0.70

osed

-0.68

ogun

-0.67

rette

-0.66

ramid

-0.65

insula

-0.65

POSITIVE LOGITS

 Wales

0.82

ford

0.79

hips

0.76

port

0.74

gar

0.72

 Bale

0.71

abouts

0.70

heed

0.69

pool

0.69

izard

0.69

Activations Density 0.039%

words related to different geographic regions, especially Wales

references to Wales and Welsh identity

No Comments

No Known Activations