INDEX

Explanations

mentions of a specific word, "Dun", in various contexts

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

>>>>>>>>

-0.71

INC

-0.69

 pains

-0.68

OHN

-0.68

xual

-0.63

NRS

-0.62

ONS

-0.62

 attachments

-0.61

 passionate

-0.60

Dys

-0.60

POSITIVE LOGITS

edin

1.14

geon

1.12

wich

1.09

querque

1.08

ogh

1.06

geons

1.05

igan

1.02

lap

1.00

ning

0.98

leys

0.97

Activations Density 0.014%

mentions of a specific word, "Dun", in various contexts

No Comments

No Known Activations