INDEX

Explanations

instances of the word "Kan" followed by numbers varying from 8 to 10

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Kan" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Ö¼

-0.77

dinand

-0.77

inals

-0.74

 Aires

-0.64

×Ļ×

-0.64

idency

-0.63

acial

-0.63

ptive

-0.62

````

-0.61

suit

-0.61

POSITIVE LOGITS

Kan

1.09

hof

1.04

igan

0.92

owa

0.91

ji

0.88

©¶æ¥µ

0.86

amel

0.84

oka

0.83

aho

0.83

ako

0.82

Activations Density 0.010%

instances of the word "Kan" followed by numbers varying from 8 to 10

mentions of the name "Kan" and its variations

No Comments

No Known Activations