INDEX

Explanations

No Explanations Found

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

 sugg

-0.82

ombo

-0.77

 Aires

-0.76

estyles

-0.69

bodied

-0.68

aurus

-0.68

ãĤ¦ãĤ¹

-0.65

 sidx

-0.62

 colourful

-0.62

score

-0.61

POSITIVE LOGITS

Ike

0.67

bil

0.66

Customer

0.65

 Reporter

0.65

 Neighbor

0.63

ctor

0.62

 Clinton

0.60

Kis

0.59

 embargo

0.58

 Geoff

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.