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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

etheless

-0.84

 lull

-0.72

 confir

-0.66

strip

-0.65

stad

-0.64

Band

-0.62

`.

-0.61

raid

-0.60

Memory

-0.60

POSITIVE LOGITS

yn

1.21

kefeller

0.78

idon

0.72

htaking

0.70

ayne

0.69

ynthesis

0.67

NOW

0.66

xit

0.65

acebook

0.65

 endings

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.