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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 prints

-0.73

bos

-0.67

sych

-0.65

 Aires

-0.64

 Limits

-0.63

lines

-0.62

 allot

-0.61

pse

-0.61

undai

-0.60

 resil

-0.60

POSITIVE LOGITS

pmwiki

0.68

atern

0.68

ãĥ¤

0.67

ESSION

0.64

 Andersen

0.64

Pok

0.63

ISSION

0.62

vous

0.62

 guiActiveUnfocused

0.61

ackle

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.