INDEX

Explanations

positive adverbs emphasizing a high degree or intensity

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_fs12288-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

12,288

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

icipated

-0.66

inarily

-0.64

idia

-0.60

 Hammond

-0.59

enance

-0.58

craft

-0.57

è¦ļéĨĴ

-0.56

Rus

-0.56

orgetown

-0.55

seless

-0.55

POSITIVE LOGITS

oths

0.79

 much

0.75

 afraid

0.73

 busy

0.73

 many

0.72

 tempting

0.71

far

0.70

leep

0.70

 guiActiveUn

0.70

ls

0.68

Activations Density 0.039%

positive adverbs emphasizing a high degree or intensity

No Comments

No Known Activations