INDEX

Explanations

phrases related to trying to do something to the best of one's ability

oai_token-act-pair · gpt-3.5-turbo

expressions indicating the degree of effort or quantity applied to an action

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

ainer

-0.80

çīĪ

-0.65

heimer

-0.63

eri

-0.63

._

-0.62

REE

-0.62

ifferent

-0.62

efe

-0.60

TION

-0.59

hler

-0.59

POSITIVE LOGITS

 possible

1.32

 practicable

0.96

 Possible

0.81

 conceivable

0.76

 feas

0.75

yip

0.74

 feasible

0.73

 imaginable

0.72

packages

0.68

comfort

0.67

Activations Density 0.056%

phrases related to trying to do something to the best of one's ability

expressions indicating the degree of effort or quantity applied to an action

No Comments

No Known Activations