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

 Attribution

-0.68

Ily

-0.67

mot

-0.65

CLASSIFIED

-0.65

 orientation

-0.62

 insecure

-0.60

GBT

-0.58

iors

-0.58

 awaited

-0.58

MOT

-0.58

POSITIVE LOGITS

ocl

0.71

pperc

0.67

alone

0.64

 bowel

0.62

rhy

0.61

owder

0.61

 diluted

0.61

TIT

0.60

Comb

0.60

thel

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.