INDEX

Explanations

phrases related to actions taken or situations observed

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

 aver

-0.66

mong

-0.65

WER

-0.64

 notwithstanding

-0.64

negie

-0.64

gart

-0.60

Aid

-0.59

arity

-0.59

never

-0.59

¶ħ

-0.59

POSITIVE LOGITS

 puberty

0.62

 Stamford

0.61

Ju

0.60

issan

0.60

 THEM

0.60

 Shanghai

0.59

 them

0.59

 Hitman

0.59

 Lanc

0.58

him

0.58

Activations Density 0.284%

phrases related to actions taken or situations observed

No Comments

No Known Activations