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

meyer

-0.67

EStreamFrame

-0.66

skirts

-0.66

 Valve

-0.66

leans

-0.66

 biscuits

-0.65

soType

-0.65

 Westminster

-0.64

nen

-0.63

onen

-0.61

POSITIVE LOGITS

eria

0.72

iasis

0.70

apy

0.68

eous

0.67

Living

0.65

igator

0.64

erie

0.62

avor

0.61

yard

0.60

enza

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.