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

abad

-0.73

amia

-0.71

actic

-0.71

ername

-0.69

ãĥ¼ãĤ¯

-0.68

avid

-0.67

othe

-0.66

amaz

-0.65

ahime

-0.65

ãĥ«

-0.64

POSITIVE LOGITS

©¶æ

0.75

 clipboard

0.69

arine

0.65

 reliant

0.63

 bunk

0.61

IPM

0.61

 upgraded

0.61

 ignores

0.61

loo

0.60

 upgrades

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.