INDEX

Explanations

locations and names

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

ahime

-0.63

 delegation

-0.59

 Parables

-0.59

viz

-0.56

aukee

-0.54

ISION

-0.53

 visualization

-0.52

okers

-0.52

 name

-0.52

 crossover

-0.51

POSITIVE LOGITS

wards

1.15

packs

1.13

packing

0.99

stab

0.98

dated

0.95

lash

0.94

fires

0.93

ward

0.91

)=(

0.89

gam

0.89

Activations Density 0.049%

locations and names

No Comments

No Known Activations