INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 rall

-0.73

mania

-0.73

 Doodle

-0.70

 SPORTS

-0.68

endon

-0.68

ãĥŃ

-0.68

ornings

-0.68

cko

-0.67

 Rhodes

-0.67

umbn

-0.66

POSITIVE LOGITS

 assass

0.72

 unsub

0.68

gency

0.65

matically

0.64

 reper

0.63

hot

0.62

 viability

0.61

 plausible

0.60

 prints

0.60

conserv

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.