INDEX

Explanations

phrases related to the usage or manipulation of resources or information

oai_token-act-pair · gpt-3.5-turbo

verbs related to the concept of "use" in various contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Stud

-0.65

llah

-0.64

Â¯Â¯Â¯Â¯

-0.64

watching

-0.64

cade

-0.63

aintain

-0.63

Enlarge

-0.62

cellence

-0.62

pleted

-0.62

rb

-0.61

POSITIVE LOGITS

fully

1.03

 coercion

0.99

 phrases

0.97

 eminent

0.91

 leverage

0.91

 intimidation

0.90

 caution

0.89

 metaphors

0.88

 discretion

0.88

 coercive

0.88

Activations Density 0.131%

phrases related to the usage or manipulation of resources or information

verbs related to the concept of "use" in various contexts

No Comments

No Known Activations