INDEX

Explanations

references to the name "Kan"

oai_token-act-pair · gpt-3.5-turbo

references to "Kan" and variations of the word

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

-0.85

 Aires

-0.77

×Ļ×

-0.69

 FIGHT

-0.68

é¾

-0.67

advertisement

-0.66

inals

-0.64

 Carbuncle

-0.63

suit

-0.62

 predictive

-0.61

POSITIVE LOGITS

hof

1.05

owa

0.98

ji

0.97

igan

0.96

ako

0.94

awa

0.93

igans

0.91

itary

0.91

amel

0.90

keye

0.89

Activations Density 0.011%

references to the name "Kan"

references to "Kan" and variations of the word

No Comments

No Known Activations