INDEX

Explanations

mentions of countries

oai_token-act-pair · gpt-3.5-turbo

mention of the country Uruguay

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

alyst

-0.81

asts

-0.78

ICS

-0.78

umbnails

-0.77

obe

-0.75

umbn

-0.74

NetMessage

-0.74

leton

-0.74

atform

-0.73

sylvania

-0.72

POSITIVE LOGITS

arez

0.94

 Uruguay

0.86

gur

0.85

recy

0.84

uran

0.83

 Urug

0.78

rig

0.75

gent

0.74

 Uzbek

0.74

endiary

0.73

Activations Density 0.030%

mentions of countries

mention of the country Uruguay

No Comments

No Known Activations