INDEX

Explanations

words related to scientific experiments and research

oai_token-act-pair · gpt-3.5-turbo

the letter "p"

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

xual

-0.86

pite

-0.84

thumbnails

-0.80

artist

-0.80

pter

-0.76

cycles

-0.75

igslist

-0.73

pmwiki

-0.73

esan

-0.72

Python

-0.72

POSITIVE LOGITS

''

0.86

 commencement

0.69

 death

0.64

 reused

0.63

 Wilkinson

0.63

 plain

0.62

 bridge

0.60

 abolished

0.60

 topic

0.60

 frontier

0.60

Activations Density 0.000%

words related to scientific experiments and research

the letter "p"

No Comments

No Known Activations