INDEX

Explanations

random text and formatting elements

oai_token-act-pair · gpt-3.5-turbo

special characters or symbols within the text

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

pse

-0.83

 behav

-0.80

 Pyth

-0.55

 behavi

-0.55

 referen

-0.54

quished

-0.53

esp

-0.52

 disadvant

-0.52

alore

-0.51

 tradem

-0.51

POSITIVE LOGITS

RPG

0.70

GH

0.69

âķĲâķĲ

0.66

Ùħ

0.64

à¨

0.64

Ø³

0.63

ãģĨ

0.62

×ª

0.62

ãĥ´ãĤ¡

0.61

Mars

0.61

Activations Density 0.215%

random text and formatting elements

special characters or symbols within the text

No Comments

No Known Activations