INDEX

Explanations

proper nouns or titles given to entities or objects

oai_token-act-pair · gpt-3.5-turbo

phrases that include the word "dubbed" or similar synonyms indicating a label or title

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ramid

-0.66

Admin

-0.64

 Admin

-0.64

UI

-0.63

 istg

-0.63

=-=-=-=-

-0.61

emic

-0.60

iment

-0.59

chedel

-0.58

=-=-

-0.58

POSITIVE LOGITS

 dubbed

0.75

fair

0.74

selves

0.74

"#

0.72

phas

0.71

geon

0.71

"@

0.70

é¾

0.68

comings

0.68

iously

0.68

Activations Density 0.022%

proper nouns or titles given to entities or objects

phrases that include the word "dubbed" or similar synonyms indicating a label or title

No Comments

No Known Activations