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

chains

-0.74

ieth

-0.73

rises

-0.72

è¦

-0.71

é¾į

-0.70

 Dunk

-0.69

fulness

-0.68

wards

-0.68

æĪ¦

-0.68

pher

-0.67

POSITIVE LOGITS

etheless

0.77

 intangible

0.70

ĸļ

0.65

IPM

0.65

 emot

0.64

 Slime

0.64

alore

0.63

krit

0.62

ebin

0.61

neb

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.