INDEX

Explanations

keywords related to technology, programming languages, and specific companies

oai_token-act-pair · gpt-3.5-turbo

references to specific terms or entities, particularly related to technology, culture, and social issues

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

-0.60

nz

-0.56

"]=>

-0.56

rss

-0.54

ived

-0.52

hh

-0.51

nos

-0.51

berra

-0.51

ync

-0.51

ãĥķãĤ©

-0.50

POSITIVE LOGITS

 itself

0.63

 herself

0.58

rul

0.54

 phenotype

0.53

 himself

0.53

 peripher

0.50

 Citadel

0.50

 tong

0.49

 elect

0.47

 Facts

0.47

Activations Density 0.851%

keywords related to technology, programming languages, and specific companies

references to specific terms or entities, particularly related to technology, culture, and social issues

No Comments

No Known Activations