INDEX

Explanations

words related to experimentation and experience

oai_token-act-pair · gpt-3.5-turbo

words related to experience and experimental processes

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

 Appalachian

-0.86

fam

-0.78

forth

-0.70

apo

-0.64

xual

-0.63

 vigil

-0.61

 Appalach

-0.60

 Sabha

-0.59

 Kinnikuman

-0.59

uyomi

-0.58

POSITIVE LOGITS

ienced

1.38

exper

1.22

iences

1.12

iments

1.09

imental

1.09

iment

1.00

 Exper

0.97

ience

0.94

iven

0.93

icion

0.87

Activations Density 0.010%

words related to experimentation and experience

words related to experience and experimental processes

No Comments

No Known Activations