INDEX

Explanations

numbers in the data

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "eight."

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

Fed

-0.70

UGE

-0.70

igation

-0.67

hammad

-0.66

rison

-0.62

reference

-0.61

 depended

-0.60

plom

-0.59

å£«

-0.58

actionDate

-0.58

POSITIVE LOGITS

eenth

1.44

een

1.43

teen

1.27

teenth

1.18

 hundred

1.00

ieth

1.00

fif

0.94

eteen

0.93

 months

0.89

aciously

0.87

Activations Density 0.012%

numbers in the data

occurrences of the word "eight."

No Comments

No Known Activations