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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Spoiler

-0.76

uph

-0.70

venants

-0.68

Runner

-0.68

onda

-0.66

OH

-0.66

okes

-0.66

KC

-0.63

!--

-0.63

};

-0.63

POSITIVE LOGITS

 towed

0.73

ancial

0.71

oscopic

0.67

lyak

0.66

aditional

0.65

icipated

0.65

 vantage

0.65

latable

0.64

iege

0.63

acterial

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.