INDEX

Explanations

words associated with measurements or statistics, particularly in health and demographic 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

 salv

-0.66

 camouflage

-0.64

 imitation

-0.63

 backfield

-0.63

 habit

-0.63

 fuss

-0.62

 troop

-0.62

 stump

-0.62

 intercepted

-0.62

 scram

-0.61

POSITIVE LOGITS

ï¸ı

1.22

 whereas

1.03

endif

0.97

except

0.92

although

0.92

yet

0.92

unless

0.91

ï¸

0.91

 Additionally

0.91

thus

0.90

Activations Density 0.259%

words associated with measurements or statistics, particularly in health and demographic contexts

No Comments

No Known Activations