INDEX

Explanations

contractions involving "does not"

oai_token-act-pair · gpt-3.5-turbo

negations or phrases indicating disagreement or denial

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Classification

-0.68

 protected

-0.65

PU

-0.64

 learning

-0.63

 Carth

-0.62

 nearest

-0.62

 Butt

-0.60

 Letter

-0.60

 elimination

-0.60

 pockets

-0.59

POSITIVE LOGITS

't

1.64

ÃŃ

1.01

Â´

0.98

etsk

0.91

uts

0.91

0.90

ates

0.90

acio

0.90

itely

0.88

eness

0.87

Activations Density 0.124%

contractions involving "does not"

negations or phrases indicating disagreement or denial

No Comments

No Known Activations