INDEX

Explanations

references to specific percentage reductions or increases in values

oai_token-act-pair · gpt-3.5-turbo

numerical data, particularly percentages and dollar amounts related to financial or health statistics

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

wait

-0.65

 ling

-0.65

eatured

-0.58

rology

-0.56

iman

-0.56

 Pengu

-0.55

ibaba

-0.55

culosis

-0.54

Reviewer

-0.54

arcity

-0.54

POSITIVE LOGITS

%.

1.00

%,

0.82

dB

0.78

%;

0.70

 thirds

0.68

 degrees

0.68

%:

0.66

 margin

0.66

IGHT

0.64

fold

0.64

Activations Density 0.143%

references to specific percentage reductions or increases in values

numerical data, particularly percentages and dollar amounts related to financial or health statistics

No Comments

No Known Activations