INDEX

Explanations

information related to food and nutrition

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

 Wink

-0.84

 Caldwell

-0.76

Jed

-0.74

assis

-0.69

owered

-0.68

Gib

-0.67

 Overt

-0.66

 Woods

-0.65

 Freem

-0.64

POSITIVE LOGITS

stuff

1.18

grain

1.10

borne

1.01

 eaten

1.00

 allergies

0.95

corn

0.94

 additives

0.92

mone

0.91

 consumption

0.89

 intake

0.89

Activations Density 0.031%

information related to food and nutrition

No Comments

No Known Activations