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

terday

-0.82

wagen

-0.80

hester

-0.79

 srfAttach

-0.77

lr

-0.76

hess

-0.75

Ford

-0.73

ounter

-0.72

gow

-0.72

opol

-0.69

POSITIVE LOGITS

 aeros

0.66

 Haitian

0.65

ide

0.61

 envision

0.61

 Hait

0.61

 quotations

0.61

 attent

0.60

 logos

0.60

ãĥ¼ãĥ³

0.60

 Earthqu

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.