INDEX

Explanations

terms related to organizational structures and roles in various contexts

oai_token-act-pair · gpt-4o-mini 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

humans

-0.65

wives

-0.62

dylib

-0.61

bows

-0.61

 Things

-0.61

ersive

-0.60

elight

-0.60

 Guys

-0.59

 railways

-0.59

kies

-0.58

POSITIVE LOGITS

 individually

1.31

 separately

1.05

 imaginable

0.91

 except

0.85

 participant

0.81

 corresponds

0.81

 ounce

0.79

 independently

0.77

 contributes

0.73

 thereafter

0.72

Activations Density 0.112%

terms related to organizational structures and roles in various contexts

No Comments

No Known Activations