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

iesta

-0.75

 Alger

-0.70

llers

-0.69

ulp

-0.68

Ł

-0.66

åĤ

-0.66

{\

-0.64

Hour

-0.64

vernment

-0.63

 Bulg

-0.63

POSITIVE LOGITS

NEC

0.73

MK

0.68

bridge

0.66

IPS

0.62

UC

0.61

PF

0.60

kens

0.60

HOT

0.59

 Kens

0.58

ipal

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.