INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

papers

-0.88

imeters

-0.83

osures

-0.76

-+-+

-0.72

RAW

-0.71

thodox

-0.69

imeter

-0.69

 Noise

-0.68

sbm

-0.68

hai

-0.67

POSITIVE LOGITS

bal

0.74

 Phill

0.72

lil

0.68

 captivity

0.66

EVA

0.64

 horr

0.64

 narc

0.62

Duration

0.60

Pv

0.60

 ........

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.