INDEX

Explanations

action words related to stimulus or reaction

oai_token-act-pair · gpt-3.5-turbo

concepts related to triggering mechanisms and stimuli

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

earchers

-0.79

 bush

-0.69

âĶĢâĶĢ

-0.66

 nudity

-0.65

sein

-0.63

sou

-0.61

 bottleneck

-0.61

 flowed

-0.59

 grease

-0.58

 crate

-0.58

POSITIVE LOGITS

gamer

0.71

 Lists

0.70

imony

0.69

 Claims

0.69

tv

0.69

areth

0.67

Instance

0.66

uli

0.64

atches

0.64

ICH

0.63

Activations Density 0.001%

action words related to stimulus or reaction

concepts related to triggering mechanisms and stimuli

No Comments

No Known Activations