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

ogie

-0.82

Reviewer

-0.81

PG

-0.73

Temperature

-0.70

storage

-0.70

GREEN

-0.69

ANA

-0.68

Yellow

-0.67

ĨĴ

-0.67

ahime

-0.66

POSITIVE LOGITS

 susp

0.68

 sugg

0.64

 Stim

0.61

 concess

0.61

 exercises

0.60

 brace

0.59

reon

0.58

 scram

0.58

Dil

0.58

 poems

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.