INDEX

Explanations

technical terms related to technology and science, specifically focusing on proteins or processes involving proteins

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

manship

-0.82

è£ħ

-0.81

INESS

-0.72

hyde

-0.71

 GOODMAN

-0.68

 Learns

-0.66

 DRAG

-0.66

 Dying

-0.65

FORE

-0.64

burning

-0.62

POSITIVE LOGITS

otyp

1.29

agonist

1.26

ocol

1.23

racted

1.22

ractor

1.21

ected

1.21

otypes

1.17

ests

1.12

agonists

1.07

ective

1.03

Activations Density 0.022%

technical terms related to technology and science, specifically focusing on proteins or processes involving proteins

No Comments

No Known Activations