INDEX

Explanations

proper nouns and names of places or entities

oai_token-act-pair · gpt-3.5-turbo

specific nouns related to people, organizations, or roles

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

Houston

-0.74

 Houston

-0.67

 amen

-0.65

yss

-0.65

 Ellison

-0.62

 TAMADRA

-0.62

 Hawkins

-0.62

asonable

-0.61

ealous

-0.61

itive

-0.60

POSITIVE LOGITS

 abroad

0.76

isine

0.75

orate

0.75

 Orchestra

0.75

iang

0.73

pora

0.70

azeera

0.69

 diplomat

0.68

eno

0.67

Äį

0.66

Activations Density 0.435%

proper nouns and names of places or entities

specific nouns related to people, organizations, or roles

No Comments

No Known Activations