INDEX

Explanations

mentions of the word "Parks."

oai_token-act-pair · gpt-3.5-turbo

references to parks and park-related entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

oldemort

-0.70

 diagnosis

-0.69

ilde

-0.63

 diagnoses

-0.62

ilies

-0.62

 dawn

-0.60

atever

-0.60

ients

-0.59

minded

-0.59

ppo

-0.59

POSITIVE LOGITS

conservancy

0.98

 Parks

0.97

park

0.84

mallow

0.84

manship

0.82

opa

0.78

peed

0.77

ville

0.77

itect

0.76

DragonMagazine

0.73

Activations Density 0.006%

mentions of the word "Parks."

references to parks and park-related entities

No Comments

No Known Activations