INDEX

Explanations

the word "Again" and variations of it

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

izons

-0.84

rament

-0.78

ocene

-0.74

ottage

-0.68

*/(

-0.65

arthed

-0.63

oes

-0.63

oreal

-0.62

oslav

-0.61

jam

-0.61

POSITIVE LOGITS

adays

0.73

 reiterate

0.72

entimes

0.70

 conclud

0.69

nces

0.69

 echoing

0.68

tical

0.68

 unsurprisingly

0.65

forth

0.65

using

0.64

Activations Density 0.022%

the word "Again" and variations of it

No Comments

No Known Activations