INDEX

Explanations

words related to opinions, attitudes, or judgments

oai_token-act-pair · gpt-3.5-turbo

terms related to perception and how people interpret actions or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Clicker

-0.65

FTWARE

-0.62

 Frenzy

-0.59

ften

-0.53

 Waves

-0.53

andel

-0.52

 Neural

-0.52

Peb

-0.51

 Sebast

-0.51

REM

-0.51

POSITIVE LOGITS

 favorably

1.12

 skept

1.09

as

1.01

 negatively

1.00

 positively

0.89

 unfairly

0.89

 differently

0.87

 harshly

0.84

tical

0.81

 derog

0.80

Activations Density 0.099%

words related to opinions, attitudes, or judgments

terms related to perception and how people interpret actions or events

No Comments

No Known Activations