INDEX

Explanations

phrases that emphasize the degree or extent of something

oai_token-act-pair · gpt-3.5-turbo

phrases indicating varying levels or degrees of a particular quality or condition

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

Negative Logits

ades

-0.90

ãĤ¼

-0.84

ets

-0.83

ade

-0.80

okes

-0.79

large

-0.78

zens

-0.77

waves

-0.76

parts

-0.76

paces

-0.74

POSITIVE LOGITS

 sophistication

1.47

 abstraction

1.34

 scrutiny

1.33

 professionalism

1.28

 detail

1.26

 specificity

1.18

 seriousness

1.16

 difficulty

1.15

 intensity

1.14

 maturity

1.14

Activations Density 0.088%

phrases that emphasize the degree or extent of something

phrases indicating varying levels or degrees of a particular quality or condition

No Known Activations