INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

jri

-0.95

liction

-0.77

iencies

-0.77

Nap

-0.76

zik

-0.75

Aid

-0.74

itiveness

-0.71

CHA

-0.70

advertisement

-0.69

TPPStreamerBot

-0.69

POSITIVE LOGITS

theless

0.71

 Gang

0.63

 roots

0.63

 leverage

0.61

 Clover

0.60

 roadside

0.59

 tract

0.58

 savings

0.57

 Pione

0.57

 Integ

0.56

Activations Density 0.000%

No Known Activations

This feature has no known activations.