INDEX

Explanations

references to numerical figures related to percentages or quantities

oai_token-act-pair · gpt-3.5-turbo

the percentage of a whole represented by the number 90

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

lay

-0.80

wan

-0.79

board

-0.75

lish

-0.73

holder

-0.73

pocket

-0.72

holding

-0.70

stand

-0.70

maid

-0.70

REDACTED

-0.69

POSITIVE LOGITS

orsi

0.90

0.86

idth

0.86

 degrees

0.84

Âº

0.82

%"

0.82

ruciating

0.80

Â°

0.80

%:

0.80

olt

0.79

Activations Density 0.042%

references to numerical figures related to percentages or quantities

the percentage of a whole represented by the number 90

No Comments

No Known Activations