INDEX

Explanations

references to actions or states of being

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

noticed

-0.77

checking

-0.76

course

-0.70

apps

-0.66

leaf

-0.64

checked

-0.64

blocks

-0.63

 Checking

-0.62

processing

-0.62

blooded

-0.61

POSITIVE LOGITS

be

1.16

 originate

1.13

 have

1.06

 derive

1.05

 embody

1.05

 contain

1.02

 belong

0.99

 comprise

0.97

 reside

0.97

 undergo

0.97

Activations Density 0.073%

references to actions or states of being

No Comments

No Known Activations