INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ersen

-0.78

sonian

-0.78

rison

-0.77

ylum

-0.77

ihad

-0.76

bett

-0.75

ynchronous

-0.75

auga

-0.74

acus

-0.73

ked

-0.73

POSITIVE LOGITS

 Turns

0.69

PID

0.67

ãĥİ

0.67

WAYS

0.66

 masks

0.66

Ny

0.66

ãĤ§

0.61

ops

0.61

EXP

0.59

 runtime

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.