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

 Bake

-0.68

 deficit

-0.64

 adore

-0.62

 mortg

-0.62

 BIOS

-0.62

 dreamed

-0.61

antha

-0.61

 NETWORK

-0.60

arth

-0.60

isode

-0.59

POSITIVE LOGITS

Sharp

0.77

elsen

0.75

 clauses

0.68

istar

0.66

icas

0.65

tl

0.63

encers

0.62

 executions

0.61

AAF

0.61

 methods

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.