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

 Shades

-0.76

 Recover

-0.73

 Deter

-0.64

 Landing

-0.64

 Gathering

-0.63

 Selected

-0.61

 Payments

-0.61

 alternate

-0.60

SS

-0.60

 Bulk

-0.60

POSITIVE LOGITS

 Guinness

0.73

issan

0.73

âĹ¼

0.70

laz

0.69

luaj

0.69

cig

0.69

itten

0.67

iceps

0.67

Downloadha

0.65

enza

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.