INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Logged

-0.56

?,

-0.54

onian

-0.53

 ãĤµ

-0.47

mouth

-0.47

 Enough

-0.47

 Operator

-0.46

Id

-0.46

erson

-0.46

 accelerator

-0.46

POSITIVE LOGITS

SPONSORED

0.64

Benz

0.60

 Tanz

0.58

merce

0.57

undai

0.57

idal

0.56

ichever

0.56

rys

0.55

kefeller

0.55

liter

0.55

Activations Density 0.000%

No Known Activations

This feature has no known activations.