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

 Titanic

-0.77

 Darling

-0.74

iley

-0.73

 Instr

-0.71

 Gaga

-0.69

istle

-0.69

cz

-0.69

ãĥ¼ãĥĨãĤ£

-0.69

Ear

-0.67

Stars

-0.67

POSITIVE LOGITS

 pacing

0.76

grown

0.72

aredevil

0.67

lov

0.67

edge

0.66

hun

0.65

aqu

0.65

 backyard

0.64

 tops

0.62

 tofu

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.