INDEX

Explanations

engineering-related terms and technical instructions

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

ĸļ

-0.89

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.81

hood

-0.75

 Dare

-0.74

 guiActiveUnfocused

-0.69

wich

-0.67

 Shutterstock

-0.67

EngineDebug

-0.66

 Territories

-0.66

naires

-0.66

POSITIVE LOGITS

rylic

1.40

oustic

1.34

robat

1.26

upuncture

1.16

osta

1.12

oust

1.11

anth

1.08

acia

1.08

ademic

1.04

rament

1.03

Activations Density 0.017%

engineering-related terms and technical instructions

No Comments

No Known Activations