INDEX

Explanations

words related to political discussions and actions

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

 âĢº

-0.70

 depends

-0.69

enery

-0.65

teness

-0.64

 varies

-0.64

zees

-0.64

emouth

-0.63

)",

-0.63

"},{"

-0.62

 =====

-0.60

POSITIVE LOGITS

 filib

0.80

 disav

0.76

 hastily

0.76

 denouncing

0.75

 scorn

0.75

 ridicule

0.73

 lobbying

0.72

 ensued

0.72

 scathing

0.71

 indignation

0.70

Activations Density 1.237%

words related to political discussions and actions

No Comments

No Known Activations