INDEX

Explanations

mentions of the word "ory"

oai_token-act-pair · gpt-3.5-turbo

words or phrases related to categories or classifications

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

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

 Archdemon

-0.69

 Rebels

-0.69

 smugglers

-0.66

 hypers

-0.61

 newsletters

-0.60

 Appalach

-0.60

 fostering

-0.60

 timelines

-0.58

 FEMA

-0.58

 fertile

-0.57

POSITIVE LOGITS

ory

1.44

ORY

1.11

ona

0.97

ories

0.91

anka

0.88

orio

0.88

orie

0.87

heastern

0.79

oya

0.77

pheus

0.75

Activations Density 0.007%

mentions of the word "ory"

words or phrases related to categories or classifications

No Comments

No Known Activations