INDEX

Explanations

specific terms related to measurements or quality evaluation, especially the term 'caliber'

oai_token-act-pair · gpt-3.5-turbo

terms related to quality or standard of excellence

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Commons

-0.64

sun

-0.62

 Pinterest

-0.62

hee

-0.62

ENA

-0.62

tree

-0.60

opian

-0.59

HI

-0.59

 trolls

-0.59

 Shack

-0.58

POSITIVE LOGITS

 caliber

4.01

 calib

3.44

caliber

2.78

 stature

1.35

 Bullets

1.18

 revolver

1.12

 ammunition

1.09

 ammo

1.05

 Colt

1.04

 cartridge

1.03

Activations Density 0.030%

specific terms related to measurements or quality evaluation, especially the term 'caliber'

terms related to quality or standard of excellence

No Comments

No Known Activations