INDEX

Explanations

references to low-quality or unwanted items or materials

oai_token-act-pair · gpt-3.5-turbo

references to "junk" in various contexts, particularly related to food and financial terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

oral

-0.77

soType

-0.75

gypt

-0.75

Æ

-0.73

arya

-0.73

staking

-0.72

zza

-0.71

cing

-0.71

tnc

-0.71

phrine

-0.69

POSITIVE LOGITS

 junk

0.93

 heap

0.85

 mail

0.79

 drawer

0.79

ãģĦ

0.75

ãģı

0.73

mail

0.71

bags

0.70

 pile

0.70

town

0.69

Activations Density 0.039%

references to low-quality or unwanted items or materials

references to "junk" in various contexts, particularly related to food and financial terms

No Comments

No Known Activations