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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Ronnie

-0.74

riot

-0.73

hani

-0.67

 incumb

-0.65

 mosqu

-0.63

 psychiat

-0.63

 Comput

-0.61

 mathemat

-0.61

 NASL

-0.61

 evid

-0.61

POSITIVE LOGITS

ulnerable

0.79

pling

0.75

edom

0.74

zai

0.70

scape

0.69

zag

0.68

amily

0.67

gaard

0.67

ILY

0.67

million

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.