INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

negie

-0.71

ibrary

-0.63

 caus

-0.62

 Providence

-0.61

Rev

-0.61

 finger

-0.60

LL

-0.60

 Dent

-0.58

 determin

-0.58

HW

-0.57

POSITIVE LOGITS

RAM

0.92

oft

0.76

Frames

0.76

camp

0.76

gran

0.71

pex

0.71

Gas

0.70

MAP

0.70

modules

0.69

 targ

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.