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

 Dialogue

-0.78

fleet

-0.73

Dialogue

-0.73

lag

-0.69

ilic

-0.67

umph

-0.67

 Governors

-0.66

Row

-0.66

icultural

-0.65

haus

-0.65

POSITIVE LOGITS

sar

0.64

 skelet

0.64

 thigh

0.62

 therap

0.61

zip

0.61

 CARE

0.60

 knee

0.59

 heaviest

0.59

 shell

0.58

 tumor

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.