INDEX

Explanations

phrases related to completion or conclusions, often involving terms like 'said' or 'done'

oai_token-act-pair · gpt-3.5-turbo

variations and repetitions of the concept of forms and conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

imens

-0.62

aminer

-0.58

ioned

-0.51

imer

-0.49

iste

-0.48

ucl

-0.48

riot

-0.46

 linkage

-0.45

 Toxic

-0.45

Ward

-0.45

POSITIVE LOGITS

1.03

0.98

0.88

terday

0.85

0.80

ï¸

0.76

ayers

0.74

AND

0.70

Ô

0.69

and

0.69

Activations Density 0.817%

phrases related to completion or conclusions, often involving terms like 'said' or 'done'

variations and repetitions of the concept of forms and conditions

No Comments

No Known Activations