INDEX

Explanations

nouns related to politics and government

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

rompt

-0.82

heid

-0.65

 Mellon

-0.65

alid

-0.64

 Chains

-0.63

vest

-0.61

moil

-0.61

 CARD

-0.60

icer

-0.60

edin

-0.58

POSITIVE LOGITS

 importantly

1.32

afa

0.99

 notably

0.98

 mornings

0.90

likely

0.83

 likely

0.82

body

0.81

 egreg

0.80

 noticeable

0.77

 observers

0.75

Activations Density 0.262%

nouns related to politics and government

No Comments

No Known Activations