INDEX

Explanations

terminology related to the military and organizations

oai_token-act-pair · gpt-3.5-turbo

references to "AF" followed by numerical values or context related to organizations or entities

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

 Kessler

-0.73

 Muller

-0.72

 Olympia

-0.71

 Boone

-0.70

 Blackwell

-0.68

 Heller

-0.66

MLG

-0.66

igators

-0.66

 Kissinger

-0.65

 lies

-0.62

POSITIVE LOGITS

RIC

1.21

rica

1.06

rican

1.05

onso

0.98

avorite

0.95

chieve

0.95

ornia

0.95

SPA

0.92

lesh

0.92

AF

0.91

Activations Density 0.004%

terminology related to the military and organizations

references to "AF" followed by numerical values or context related to organizations or entities

No Comments

No Known Activations