INDEX

Explanations

words related to filling gaps or voids

oai_token-act-pair · gpt-3.5-turbo

terms associated with missing elements or deficiencies

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

ETHOD

-0.77

ITNESS

-0.70

 approves

-0.62

uru

-0.59

anmar

-0.59

Compare

-0.59

arettes

-0.59

 consolation

-0.58

independent

-0.57

 livest

-0.56

POSITIVE LOGITS

 created

0.87

 plag

0.87

 caused

0.85

 arising

0.81

 surrounding

0.80

 inherited

0.78

 inherent

0.76

 generated

0.76

 incurred

0.76

 regarding

0.75

Activations Density 0.262%

words related to filling gaps or voids

terms associated with missing elements or deficiencies

No Comments

No Known Activations