INDEX

Explanations

mentions of specific technical terms or product names

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter 'a'

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

-0.66

 pige

-0.62

most

-0.61

lasses

-0.59

sed

-0.58

bart

-0.57

aer

-0.57

 curfew

-0.57

DON

-0.57

sin

-0.56

POSITIVE LOGITS

Ã±

1.21

veland

1.11

Ã§

1.10

Ã±a

1.03

Ã°

1.02

emon

1.01

ÅŁ

1.00

ples

1.00

issance

0.99

vel

0.98

Activations Density 0.079%

mentions of specific technical terms or product names

occurrences of the letter 'a'

No Comments

No Known Activations