INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

mania

-0.72

 Stru

-0.71

 Rouge

-0.67

_>

-0.67

Ner

-0.64

 embry

-0.62

imar

-0.62

 Mania

-0.62

 mismatch

-0.61

 favour

-0.60

POSITIVE LOGITS

ctory

0.75

romptu

0.72

asionally

0.71

ocated

0.70

ational

0.69

ocations

0.67

rued

0.66

ealous

0.66

itional

0.65

alties

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.