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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ether

-0.69

 Onion

-0.67

Lib

-0.65

 Guinea

-0.64

sword

-0.64

 Instruments

-0.63

eon

-0.62

ldom

-0.62

pex

-0.61

hered

-0.61

POSITIVE LOGITS

 rall

0.75

Pg

0.69

intendent

0.67

acci

0.66

 sqor

0.64

 annex

0.64

orer

0.62

 overdose

0.61

arios

0.61

CHA

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.