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

psy

-0.82

nz

-0.79

vic

-0.73

stice

-0.69

ago

-0.67

trop

-0.66

ube

-0.66

css

-0.66

wx

-0.65

azz

-0.65

POSITIVE LOGITS

he

0.72

ortunately

0.65

quartered

0.64

sir

0.61

ß

0.61

SHE

0.61

luster

0.60

 recourse

0.60

 THEY

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.