INDEX

Explanations

phrases indicating confusion or lack of logical sense

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a lack of clarity or understanding

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

gypt

-0.88

alus

-0.69

 ensu

-0.68

hra

-0.67

alla

-0.63

 nowhere

-0.63

alian

-0.60

 WATCH

-0.60

ean

-0.59

 outnumbered

-0.58

POSITIVE LOGITS

 sense

1.22

sense

1.14

 bones

1.11

 difference

1.05

 assumptions

0.94

 allowance

0.94

 Sense

0.93

 allowances

0.91

 distinction

0.90

 noise

0.87

Activations Density 0.042%

phrases indicating confusion or lack of logical sense

phrases indicating a lack of clarity or understanding

No Comments

No Known Activations