INDEX

Explanations

terms related to viewpoints or perspectives

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

pload

-0.74

rium

-0.68

esc

-0.68

arus

-0.68

redo

-0.65

urated

-0.65

ramids

-0.64

strom

-0.63

 Pirate

-0.63

olesc

-0.62

POSITIVE LOGITS

 standpoint

0.89

 anyway

0.75

 perspective

0.71

 merits

0.68

 obviously

0.66

wcsstore

0.64

 meaning

0.64

 anyways

0.64

0.62

 purely

0.62

Activations Density 0.164%

terms related to viewpoints or perspectives

No Comments

No Known Activations