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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uan

-0.71

Arthur

-0.70

Tu

-0.66

Lear

-0.65

zu

-0.65

yu

-0.64

anda

-0.63

zai

-0.63

weapon

-0.62

ccording

-0.62

POSITIVE LOGITS

ãĤ©

0.94

rament

0.74

ULAR

0.71

FORE

0.69

SOM

0.69

 RAID

0.67

 externalToEVAOnly

0.67

irgin

0.67

stone

0.66

DRM

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.