INDEX

Explanations

technical terms related to computer hardware components and programming

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

witch

-0.80

arians

-0.78

aughs

-0.73

ibl

-0.72

=\"

-0.71

olph

-0.71

conom

-0.70

abe

-0.70

ypes

-0.70

osuke

-0.69

POSITIVE LOGITS

 vicinity

1.58

 midst

1.53

 meantime

1.25

 equation

1.14

 middle

1.13

 bloodstream

1.12

 confines

1.11

 absence

1.09

 periphery

1.07

 guise

1.05

Activations Density 0.258%

technical terms related to computer hardware components and programming

No Comments

No Known Activations