INDEX

Explanations

phrases related to sizes and measurements

oai_token-act-pair · gpt-3.5-turbo

references to size comparisons or measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ãĥŃ

-0.79

oner

-0.71

rov

-0.71

vic

-0.68

Activ

-0.67

gencies

-0.66

wrong

-0.65

arak

-0.65

impro

-0.65

itions

-0.64

POSITIVE LOGITS

 postage

1.03

 Everest

0.94

 Rhode

0.93

 Manhattan

0.88

 Jupiter

0.88

 Delaware

0.83

 elephant

0.82

 iceberg

0.82

 Hiroshima

0.81

 golf

0.81

Activations Density 0.186%

phrases related to sizes and measurements

references to size comparisons or measurements

No Comments

No Known Activations