INDEX

Explanations

mentions of locations or regions, specifically provinces

oai_token-act-pair · gpt-3.5-turbo

references to a specific geographic region or administrative division

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

tle

-0.91

artz

-0.81

ãĥīãĥ©

-0.80

story

-0.75

chel

-0.75

kus

-0.75

gers

-0.72

visory

-0.72

Sym

-0.71

noon

-0.71

POSITIVE LOGITS

 province

0.92

naire

0.87

wide

0.85

 provinces

0.83

 sovere

0.79

 unemploy

0.78

 annexed

0.76

ality

0.76

 countryside

0.73

 legislatures

0.71

Activations Density 0.012%

mentions of locations or regions, specifically provinces

references to a specific geographic region or administrative division

No Comments

No Known Activations