INDEX

Explanations

references to nuclear programs and related terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

torch.float32

Hook Point

blocks.8.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

steen

-0.80

tyard

-0.75

Ort

-0.74

 Homeless

-0.72

Completed

-0.71

alach

-0.71

Preview

-0.69

Ń·

-0.69

MpServer

-0.69

 Gerrard

-0.69

POSITIVE LOGITS

 weapons

1.09

 missile

1.08

 warheads

1.04

 missiles

1.00

 nuclear

0.97

 disarm

0.97

 bomb

0.96

weapons

0.94

 arsenal

0.94

 bombs

0.93

Activations Density 0.015%

references to nuclear programs and related terminology

No Comments

No Known Activations