INDEX

Explanations

syntax elements like code comments and dots before newlines

oai_token-act-pair · gpt-3.5-turbo

ellipses and unfinished sentences or thoughts

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

ãĥ¼ãĥĨãĤ£

-0.92

uers

-0.76

 Galile

-0.70

ratulations

-0.69

ously

-0.66

 aven

-0.65

 Stall

-0.64

 Krug

-0.64

 slope

-0.63

 slopes

-0.62

POSITIVE LOGITS

etc

1.04

walking

0.87

please

0.83

where

0.81

 fixme

0.81

ordered

0.80

SER

0.77

cum

0.76

ser

0.76

Activations Density 0.011%

syntax elements like code comments and dots before newlines

ellipses and unfinished sentences or thoughts

No Comments

No Known Activations