INDEX

Explanations

No Explanations Found

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

âķĲ

-0.69

)].

-0.66

Ùĩ

-0.61

 externalToEVAOnly

-0.59

ļé

-0.59

Ð´

-0.58

ysis

-0.57

¿½

-0.57

ãĤ¼ãĤ¦ãĤ¹

-0.57

ebin

-0.57

POSITIVE LOGITS

↵

1.41

↵↵

0.92

<|endoftext|>

0.91

 Unlike

0.66

htaking

0.63

Story

0.62

 Arri

0.61

 Instead

0.60

ramid

0.57

culosis

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.