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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ONSORED

-0.80

 Recover

-0.70

Cu

-0.67

Pil

-0.67

ìĿ

-0.65

Columb

-0.65

-0.63

sic

-0.63

erity

-0.63

Quote

-0.63

POSITIVE LOGITS

 fences

0.79

 rods

0.77

 resil

0.70

ounters

0.68

ĵĺ

0.66

gad

0.64

iscons

0.64

sembly

0.64

 Bunker

0.62

 misdem

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.