INDEX

Explanations

phrases related to numerical rankings or positions

oai_token-act-pair · gpt-3.5-turbo

references to rankings and significant positions in sports or competitive contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

utic

-0.74

sym

-0.70

ventions

-0.68

ãĤ¹ãĥĪ

-0.67

adr

-0.67

apters

-0.66

uras

-0.65

-0.64

ournals

-0.63

ework

-0.63

POSITIVE LOGITS

 contender

0.99

 ranked

0.97

 seed

0.92

 spot

0.92

 ranking

0.89

 overall

0.87

 rated

0.86

 earners

0.85

 priority

0.81

ranked

0.77

Activations Density 0.045%

phrases related to numerical rankings or positions

references to rankings and significant positions in sports or competitive contexts

No Comments

No Known Activations