INDEX

Explanations

words related to food and ingredients or technical terms related to technology

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

 thereby

-0.87

ãĥ´

-0.76

owment

-0.75

Allows

-0.75

soever

-0.74

ãĤ¢ãĥ«

-0.74

rn

-0.74

hire

-0.74

rade

-0.74

since

-0.72

POSITIVE LOGITS

 basics

1.30

 hardest

1.20

 obligatory

1.14

 simplest

1.13

 easiest

1.04

 biggest

1.03

 Basics

1.02

 final

1.01

 tricky

0.99

 idea

0.98

Activations Density 0.341%

words related to food and ingredients or technical terms related to technology

No Comments

No Known Activations