INDEX

Explanations

phrases related to improvement and performing well

oai_token-act-pair · gpt-3.5-turbo

evaluations of performance or quality, particularly related to improvement and doing well

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

jected

-0.72

urated

-0.69

Tru

-0.68

ãĤ¦ãĤ¹

-0.68

 severed

-0.65

 Personality

-0.64

Tav

-0.63

 Cutter

-0.63

 cracked

-0.63

gerald

-0.63

POSITIVE LOGITS

job

0.79

 grunt

0.75

 chores

0.72

 entreprene

0.71

 offline

0.70

 homework

0.70

FUL

0.70

 repay

0.67

 injustice

0.66

 enrich

0.66

Activations Density 0.066%

phrases related to improvement and performing well

evaluations of performance or quality, particularly related to improvement and doing well

No Comments

No Known Activations