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

 Organization

-0.77

 Inspection

-0.71

 Organisation

-0.69

ãĤ´ãĥ³

-0.66

 cracks

-0.64

Aid

-0.64

ureau

-0.60

 Decoder

-0.59

 Skills

-0.59

yo

-0.59

POSITIVE LOGITS

sold

0.69

cel

0.66

cov

0.65

 bounds

0.65

waters

0.64

pees

0.63

hedon

0.63

ravel

0.62

awa

0.62

MLA

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.