INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

iple

-0.84

 commissions

-0.74

larg

-0.71

inces

-0.70

iasis

-0.68

atform

-0.66

ince

-0.66

 discontin

-0.65

iza

-0.65

escal

-0.65

POSITIVE LOGITS

Unknown

0.72

kson

0.70

ÙĴ

0.68

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

0.66

OSH

0.62

udic

0.62

herer

0.61

 incidentally

0.61

WER

0.61

 beforehand

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.