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

 unsolved

-0.72

-|

-0.69

 ï¿½ï¿½ï¿½ï¿½

-0.67

 Anarchy

-0.62

{*

-0.62

 trending

-0.62

 spirited

-0.61

 Atkinson

-0.61

 dots

-0.61

 Rudd

-0.60

POSITIVE LOGITS

natureconservancy

0.84

wake

0.79

SPONSORED

0.78

mind

0.72

fell

0.71

cedes

0.70

joy

0.70

Shares

0.68

photos

0.67

fort

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.