INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ngth

-0.77

Pokemon

-0.76

 Pokemon

-0.75

 Elys

-0.73

etheless

-0.73

theless

-0.70

 Hispan

-0.70

ricular

-0.68

esity

-0.67

TPPStreamerBot

-0.67

POSITIVE LOGITS

hips

0.75

 expire

0.72

aver

0.69

 counters

0.69

kus

0.68

aru

0.67

Export

0.66

pring

0.64

 avert

0.64

urat

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.