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

ngth

-0.93

 eleph

-0.71

icit

-0.69

licks

-0.68

hemer

-0.64

aven

-0.64

 Agents

-0.64

 traffickers

-0.62

Ñĭ

-0.62

adena

-0.60

POSITIVE LOGITS

neau

0.70

saw

0.68

ãĥĺ

0.66

surv

0.65

rg

0.65

Tile

0.64

ORPG

0.64

Shock

0.63

Discuss

0.60

 deduct

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.