INDEX

Explanations

phrases related to global concepts or entities

oai_token-act-pair · gpt-3.5-turbo

references to the word "globe" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

paren

-0.82

Klu

-0.79

Pru

-0.78

ivation

-0.76

ivas

-0.76

Ô

-0.72

alg

-0.72

CHAT

-0.71

Sequ

-0.70

ABE

-0.68

POSITIVE LOGITS

 globe

0.93

ophone

0.87

Rim

0.76

 gazing

0.76

 spread

0.73

locked

0.73

 lobe

0.72

 Reach

0.72

Ħ¢

0.72

cies

0.70

Activations Density 0.017%

phrases related to global concepts or entities

references to the word "globe" in various contexts

No Comments

No Known Activations