INDEX

Explanations

phrases related to specific geographic locations or events

oai_token-act-pair · gpt-3.5-turbo

the letter "ó" in various contexts

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

CST

-0.70

acle

-0.68

growth

-0.65

ory

-0.65

pread

-0.64

achine

-0.63

 Secretary

-0.63

GT

-0.61

 Chattanooga

-0.61

ories

-0.61

POSITIVE LOGITS

Ã³

1.12

Ãį

1.05

rien

1.00

vernment

1.00

ÃįÃį

0.97

Ã°

0.89

ÃŃa

0.89

ÃŃ

0.88

 corrid

0.85

Ã³n

0.83

Activations Density 0.006%

phrases related to specific geographic locations or events

the letter "ó" in various contexts

No Comments

No Known Activations