INDEX

Explanations

phrases indicating similarity or equivalence

oai_token-act-pair · gpt-3.5-turbo

phrases that include the word "as."

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

YD

-0.74

Whe

-0.67

uld

-0.66

ople

-0.66

Plex

-0.65

LA

-0.65

Kent

-0.64

endale

-0.64

loe

-0.64

uned

-0.63

POSITIVE LOGITS

 usual

0.80

pects

0.79

 ours

0.78

 regards

0.75

ylum

0.75

 advertised

0.74

bestos

0.72

 follows

0.70

 opposed

0.69

ocial

0.66

Activations Density 0.042%

phrases indicating similarity or equivalence

phrases that include the word "as."

No Comments

No Known Activations