INDEX

Explanations

references to behind-the-scenes events or activities

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

mington

-0.89

arcity

-0.89

haps

-0.82

liam

-0.82

ibaba

-0.78

adesh

-0.78

nesota

-0.77

heit

-0.76

renheit

-0.76

ulia

-0.76

POSITIVE LOGITS

 workings

1.05

 filming

0.81

 scenes

0.79

 dealings

0.78

 plotting

0.77

 mach

0.76

 development

0.76

ops

0.75

 insider

0.75

 briefings

0.74

Activations Density 0.040%

references to behind-the-scenes events or activities

No Comments

No Known Activations