INDEX

Explanations

proper nouns related to "Wilson"

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Wilson."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

awaru

-0.76

plate

-0.76

aft

-0.71

minded

-0.69

umen

-0.69

ousse

-0.68

otten

-0.67

Rated

-0.67

URI

-0.67

Lex

-0.66

POSITIVE LOGITS

 Chandler

0.80

ston

0.79

sey

0.79

 Wilson

0.78

 Strait

0.72

 Norton

0.71

 Everett

0.71

 Kenobi

0.71

ufact

0.70

ville

0.70

Activations Density 0.029%

proper nouns related to "Wilson"

mentions of the name "Wilson."

No Comments

No Known Activations