INDEX

Explanations

text related to programming concepts, specifically on defining and invoking functions and handling parameters

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Hutch

-0.79

odcast

-0.76

 Millennials

-0.75

brand

-0.75

 whistlebl

-0.73

 kitchens

-0.72

gaming

-0.71

smoking

-0.71

 resurg

-0.71

fare

-0.70

POSITIVE LOGITS

 integers

1.45

 parameter

1.38

 recursive

1.37

 integer

1.36

 interpol

1.33

 tuple

1.32

 initialization

1.31

 lookup

1.29

 boolean

1.26

 variables

1.26

Activations Density 3.475%

text related to programming concepts, specifically on defining and invoking functions and handling parameters

No Comments

No Known Activations