INDEX

Explanations

terms related to the nervous system, such as nerve and impulses

oai_token-act-pair · gpt-3.5-turbo

references to nerves and their functions or conditions

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

laws

-0.84

advertisement

-0.73

ioch

-0.73

atchewan

-0.69

»Ĵ

-0.68

ģĸ

-0.68

night

-0.68

arta

-0.66

 Teresa

-0.66

ALK

-0.65

POSITIVE LOGITS

 nerve

1.72

 nerves

1.14

 endings

0.97

 stim

0.88

 fibre

0.87

 pier

0.86

 tiss

0.84

fir

0.84

 fibers

0.83

 marrow

0.83

Activations Density 0.007%

terms related to the nervous system, such as nerve and impulses

references to nerves and their functions or conditions

No Comments

No Known Activations