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

llo

-0.74

ajo

-0.72

 Throne

-0.71

Fly

-0.71

patch

-0.68

 Base

-0.65

 Bench

-0.64

 Controls

-0.61

 Deploy

-0.60

 turrets

-0.60

POSITIVE LOGITS

ingen

0.81

staking

0.76

iety

0.76

ieties

0.74

inion

0.72

 unfocusedRange

0.70

rency

0.69

omas

0.68

theless

0.67

thood

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.