INDEX

Explanations

proper nouns related to politics and individuals

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals, particularly those named "Mam" and "Mikhail."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

kson

-0.77

Â¯Â¯Â¯Â¯

-0.75

ä½ľ

-0.73

Minecraft

-0.68

 Falk

-0.63

backer

-0.62

 trough

-0.60

ying

-0.60

Drug

-0.60

Syd

-0.60

POSITIVE LOGITS

Mam

0.87

amia

0.84

onde

0.83

ueller

0.80

aj

0.80

umin

0.80

lu

0.79

aret

0.78

noon

0.76

oufl

0.74

Activations Density 0.016%

proper nouns related to politics and individuals

references to specific individuals, particularly those named "Mam" and "Mikhail."

No Comments

No Known Activations