INDEX

Explanations

phrases or words indicating denial, negation, or contradiction

oai_token-act-pair · gpt-3.5-turbo

phrases that emphasize limitations or negations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ighting

-0.70

icion

-0.70

iets

-0.65

inventoryQuantity

-0.64

Stars

-0.63

 Cosponsors

-0.61

atures

-0.61

iard

-0.60

SI

-0.59

inav

-0.59

POSITIVE LOGITS

 indistinguishable

0.87

 synonymous

0.82

 confined

0.82

 mistaken

0.81

 regarded

0.79

 obliged

0.79

 extinguished

0.78

 compelled

0.78

 excluded

0.77

 subjected

0.76

Activations Density 0.204%

phrases or words indicating denial, negation, or contradiction

phrases that emphasize limitations or negations

No Comments

No Known Activations