INDEX

Explanations

terms or phrases that are considered to be relevant or useful based on collective experience

oai_token-act-pair · gpt-3.5-turbo

references to relevance or important information

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

rette

-0.88

kay

-0.76

boat

-0.75

boats

-0.74

rie

-0.73

anus

-0.73

rug

-0.71

jin

-0.71

uber

-0.70

cker

-0.70

POSITIVE LOGITS

 material

0.73

 contextual

0.73

="#

0.70

 newsp

0.70

ioned

0.69

 citations

0.68

 ingredient

0.68

ively

0.68

é¾įå¥ĳå£«

0.67

 information

0.67

Activations Density 0.026%

terms or phrases that are considered to be relevant or useful based on collective experience

references to relevance or important information

No Comments

No Known Activations