INDEX

Explanations

locations or proper nouns related to various places

oai_token-act-pair · gpt-3.5-turbo

specific named entities or titles, particularly those related to individuals or places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ipel

-0.72

 flakes

-0.71

 lows

-0.71

bog

-0.67

 chips

-0.65

theless

-0.63

 watts

-0.61

 hills

-0.61

 winds

-0.61

 thrott

-0.61

POSITIVE LOGITS

arde

0.80

kai

0.79

cart

0.75

usal

0.74

ibur

0.73

Corp

0.73

club

0.72

ĨĴ

0.72

ciation

0.70

imar

0.70

Activations Density 0.000%

locations or proper nouns related to various places

specific named entities or titles, particularly those related to individuals or places

No Comments

No Known Activations