INDEX

Explanations

companion-related words and actions

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

mberg

-0.83

UGE

-0.72

aji

-0.63

 Bulgar

-0.63

emi

-0.62

 Reloaded

-0.62

Lenin

-0.62

stem

-0.61

ihil

-0.61

uliffe

-0.60

POSITIVE LOGITS

 companions

0.93

hip

0.81

 companion

0.80

 thereto

0.77

ively

0.75

lihood

0.72

heses

0.72

hesis

0.72

lies

0.72

mented

0.71

Activations Density 0.021%

companion-related words and actions

No Comments

No Known Activations