INDEX

Explanations

proper nouns and technical terms

oai_token-act-pair · gpt-3.5-turbo

specific medical or health-related terms

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

drawn

-0.63

abre

-0.55

 CONC

-0.54

 Airbnb

-0.52

 sweep

-0.52

Gab

-0.51

 leng

-0.51

expensive

-0.50

 multim

-0.49

behind

-0.49

POSITIVE LOGITS

 Flavoring

0.67

mpire

0.64

UID

0.61

ASON

0.59

Pure

0.58

RPM

0.57

soType

0.56

zik

0.55

acle

0.55

mare

0.55

Activations Density 0.761%

proper nouns and technical terms

specific medical or health-related terms

No Comments

No Known Activations