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

rooms

-0.69

room

-0.65

viz

-0.65

~~~~

-0.65

wx

-0.65

sab

-0.65

 barg

-0.61

Î¿

-0.59

plates

-0.59

################################

-0.58

POSITIVE LOGITS

teness

0.77

usky

0.75

anan

0.75

tones

0.74

 guiActiveUn

0.73

phrine

0.72

atisf

0.71

iasm

0.70

ulner

0.69

 bidder

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.