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

olate

-0.65

CI

-0.62

cfg

-0.62

 concerned

-0.61

serial

-0.60

LAT

-0.59

UCS

-0.58

 Orient

-0.58

spec

-0.58

ANGE

-0.58

POSITIVE LOGITS

jriwal

0.86

hens

0.77

 "$:/

0.74

egal

0.73

heid

0.73

iary

0.72

oise

0.70

lasses

0.68

uo

0.67

enos

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.