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

inventoryQuantity

-0.80

 confir

-0.76

hower

-0.73

ividual

-0.68

cknowled

-0.64

lished

-0.64

veyard

-0.63

Chain

-0.62

 leakage

-0.62

 inhibitors

-0.61

POSITIVE LOGITS

–

0.77

[/

0.75

âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦

0.74

 âĢ¦"

0.73

 âĢ¦

0.71

yu

0.70

Brow

0.69

âĢ¦"

0.68

âĢ¦

0.66

oric

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.