INDEX

Explanations

phrases indicating physical or metaphorical proximity or advancement

oai_token-act-pair · gpt-3.5-turbo

phrases indicating proximity or nearing a goal

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

GW

-0.75

iasis

-0.72

å£«

-0.72

lee

-0.71

OCK

-0.71

wine

-0.69

NZ

-0.68

Smith

-0.68

AIN

-0.67

anity

-0.66

POSITIVE LOGITS

than

1.03

 than

1.01

 scrutiny

0.91

 Than

0.89

 proximity

0.89

 thereto

0.84

 examination

0.84

 "$:/

0.83

 closer

0.81

 confid

0.80

Activations Density 0.015%

phrases indicating physical or metaphorical proximity or advancement

phrases indicating proximity or nearing a goal

No Comments

No Known Activations