INDEX

Explanations

phrases related to obtaining or providing as much information or support as possible

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the need to maximize quantity or quantity-related expressions

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

eri

-0.71

ainer

-0.69

onomy

-0.68

heimer

-0.68

UTH

-0.68

EStream

-0.67

yne

-0.65

hler

-0.65

cv

-0.64

ICA

-0.64

POSITIVE LOGITS

 possible

1.07

 practicable

0.97

pects

0.73

 usual

0.72

 cheaply

0.71

 feas

0.70

catch

0.68

 Possible

0.66

lihood

0.65

perse

0.64

Activations Density 0.056%

phrases related to obtaining or providing as much information or support as possible

phrases indicating the need to maximize quantity or quantity-related expressions

No Comments

No Known Activations