INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ften

-0.90

 millenn

-0.81

 Democr

-0.81

ebin

-0.78

EStream

-0.76

vae

-0.75

ogun

-0.75

renheit

-0.73

 Canaver

-0.73

odox

-0.72

POSITIVE LOGITS

 structures

0.66

urated

0.65

 structure

0.62

panic

0.62

="#

0.61

 healthcare

0.59

 infrastructure

0.59

 Indigenous

0.59

ESL

0.59

 eclipse

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.