INDEX

Explanations

mentions of specific locations, especially related to events or news

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

arist

-0.80

naire

-0.78

abet

-0.77

arse

-0.75

DonaldTrump

-0.73

OHN

-0.73

toggle

-0.73

ischer

-0.72

ayne

-0.70

osit

-0.69

POSITIVE LOGITS

 Harbour

1.19

 Morning

0.98

 Lumpur

0.98

 Opera

0.94

 Harbor

0.92

 Wand

0.82

 harbour

0.78

FC

0.72

 suburbs

0.70

 Airport

0.70

Activations Density 0.016%

mentions of specific locations, especially related to events or news

No Comments

No Known Activations