INDEX

Explanations

common phrases related to academic, technical, and scientific topics, such as data analysis, classification, CPU specifications, and more

oai_token-act-pair · gpt-3.5-turbo

keywords associated with various social, cultural, and scientific themes

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

NEY

-0.63

aeda

-0.59

 Sparks

-0.56

enegger

-0.56

sembly

-0.55

utterstock

-0.55

amins

-0.55

azo

-0.53

 Godd

-0.53

bard

-0.52

POSITIVE LOGITS

worthiness

0.62

 coefficients

0.61

 interpol

0.60

 indices

0.57

 expenditure

0.56

 explanatory

0.56

vs

0.53

 contexts

0.53

âĢĲ

0.53

 Mathematics

0.52

Activations Density 1.229%

common phrases related to academic, technical, and scientific topics, such as data analysis, classification, CPU specifications, and more

keywords associated with various social, cultural, and scientific themes

No Comments

No Known Activations