INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

abama

-0.73

agents

-0.70

guided

-0.68

 converge

-0.67

imentary

-0.66

aber

-0.64

ourses

-0.63

fine

-0.63

theless

-0.62

orrect

-0.62

POSITIVE LOGITS

 Accessed

0.73

WS

0.65

cats

0.65

 envy

0.65

 Liberation

0.62

 brow

0.60

 pride

0.58

it

0.58

 scen

0.58

OWN

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.