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

 Credits

-0.73

flix

-0.70

uria

-0.69

 Classification

-0.67

usage

-0.65

mk

-0.62

 Stall

-0.62

orian

-0.62

osponsors

-0.61

draw

-0.60

POSITIVE LOGITS

 again

0.84

="#

0.79

Ļ

0.76

NING

0.74

Â·Â·

0.74

 anew

0.73

 Qiao

0.72

=""

0.70

 Again

0.70

ogh

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.