INDEX

Explanations

phrases related to winning and victory

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "won" and related terms indicating victory or achievement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

heter

-0.69

erity

-0.64

appa

-0.63

erous

-0.62

 scanner

-0.62

bian

-0.60

resa

-0.60

gypt

-0.60

tar

-0.60

 protr

-0.60

POSITIVE LOGITS

nings

0.94

't

0.91

ners

0.91

ced

0.82

 streak

0.77

MVP

0.77

 championships

0.76

throp

0.76

cest

0.72

 Rookie

0.72

Activations Density 0.051%

phrases related to winning and victory

occurrences of the word "won" and related terms indicating victory or achievement

No Comments

No Known Activations