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

abe

-0.64

 VIDEOS

-0.61

JC

-0.59

 spor

-0.59

EFF

-0.57

ãĥ¼ãĤ¯

-0.57

 organize

-0.56

 amuse

-0.56

 displacement

-0.55

 briefly

-0.55

POSITIVE LOGITS

âĵĺ

0.78

sylv

0.76

ulhu

0.75

ynthesis

0.70

 Reincarn

0.67

)].

0.67

nect

0.67

bang

0.65

sky

0.65

pecting

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.