INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Reloaded

-0.86

ppo

-0.82

pert

-0.78

phalt

-0.77

 millenn

-0.74

 Akron

-0.69

ooked

-0.69

von

-0.67

aido

-0.67

Pacific

-0.66

POSITIVE LOGITS

 è£ıè

0.67

{"

0.64

.</

0.64

ancial

0.63

\"

0.62

TAIN

0.62

 caster

0.62

 Ally

0.61

{*

0.60

ictive

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.