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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 sympt

-0.67

alys

-0.65

iji

-0.64

itters

-0.62

acca

-0.61

 detrim

-0.60

 detriment

-0.58

Ĭ±

-0.58

Sphere

-0.56

 inflamm

-0.56

POSITIVE LOGITS

↵

1.13

<|endoftext|>

0.78

isSpecialOrderable

0.73

↵↵

0.72

 Finally

0.71

pmwiki

0.63

 Sharing

0.61

 Dear

0.61

 Godd

0.61

qv

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.