INDEX

Explanations

terms related to making a differentiation or a distinction

oai_token-act-pair · gpt-3.5-turbo

expressions relating to the concept of distinction or differentiation

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

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

annis

-0.78

vae

-0.72

onz

-0.68

rollers

-0.66

ctic

-0.64

otin

-0.62

 Printed

-0.62

 Polo

-0.62

psc

-0.60

odes

-0.59

POSITIVE LOGITS

naire

0.95

 distinction

0.95

 distinctions

0.95

abl

0.87

otomy

0.86

inctions

0.82

ively

0.78

 warr

0.77

iple

0.76

alities

0.75

Activations Density 0.014%

terms related to making a differentiation or a distinction

expressions relating to the concept of distinction or differentiation

No Comments

No Known Activations