INDEX

Explanations

words related to geopolitical entities

oai_token-act-pair · gpt-3.5-turbo

specific identifiers or codes, possibly related to entities or classifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Rand

-0.74

 Paste

-0.73

TPP

-0.71

MER

-0.71

Sad

-0.71

Auth

-0.70

Reader

-0.69

Pand

-0.69

Trade

-0.69

 Beaut

-0.68

POSITIVE LOGITS

tsy

0.78

anson

0.74

achev

0.74

 Yugoslavia

0.72

worldly

0.71

hern

0.71

>[

0.70

 surround

0.65

poses

0.65

isphere

0.64

Activations Density 0.000%

words related to geopolitical entities

specific identifiers or codes, possibly related to entities or classifications

No Comments

No Known Activations