INDEX

Explanations

mentions of engines and related concepts

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

laus

-0.92

istani

-0.84

atana

-0.81

icho

-0.80

 Childhood

-0.76

ties

-0.76

lyak

-0.74

 Jagu

-0.74

ities

-0.74

onyms

-0.72

POSITIVE LOGITS

 powering

0.98

 compartment

0.89

bay

0.83

 braking

0.83

 powered

0.79

 malfunction

0.76

ugal

0.76

 horsepower

0.74

 ignition

0.73

 combustion

0.72

Activations Density 0.034%

mentions of engines and related concepts

No Comments

No Known Activations