INDEX

Explanations

terms related to military operations and wildlife observations

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

ãĤ´ãĥ³

-0.67

ribes

-0.63

yss

-0.62

rontal

-0.61

Ded

-0.59

ilts

-0.57

 Earn

-0.55

ilogy

-0.55

reditary

-0.54

 nutshell

-0.54

POSITIVE LOGITS

 lurking

1.33

 invading

1.00

 encro

0.99

 roaming

0.94

 trespass

0.93

 emanating

0.92

 tresp

0.88

 infiltr

0.87

 occupying

0.87

 wandering

0.86

Activations Density 0.354%

terms related to military operations and wildlife observations

No Comments

No Known Activations