INDEX

Explanations

specific nouns related to categories or lists

oai_token-act-pair · gpt-3.5-turbo

specific nouns related to entities, locations, and classifications

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

construct

-0.66

pter

-0.65

lda

-0.60

EngineDebug

-0.59

è¡

-0.57

tenance

-0.57

atorium

-0.56

bass

-0.56

sonian

-0.56

itiveness

-0.55

POSITIVE LOGITS

 vying

0.88

eries

0.77

 involved

0.74

 grouped

0.74

 eligible

0.73

 exempted

0.72

 offered

0.71

 whose

0.70

 besides

0.68

 actively

0.67

Activations Density 0.438%

specific nouns related to categories or lists

specific nouns related to entities, locations, and classifications

No Comments

No Known Activations