INDEX

Explanations

mentions of specific geographical locations or entities

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to people, places, and organizations

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

advertisement

-0.71

 sqor

-0.67

utterstock

-0.67

ramid

-0.65

EStreamFrame

-0.62

earance

-0.62

chest

-0.62

;;;;;;;;;;;;

-0.61

uyomi

-0.61

bilt

-0.61

POSITIVE LOGITS

 hadn

0.80

 forgot

0.79

 could

0.75

 lacks

0.74

 lacked

0.71

's

0.70

did

0.70

 would

0.67

 intends

0.66

 existed

0.66

Activations Density 0.419%

mentions of specific geographical locations or entities

proper nouns related to people, places, and organizations

No Comments

No Known Activations