INDEX

Explanations

formatted text samples within the token

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_fs6144-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_6144

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

6,144

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

 enjoyment

-0.66

 healed

-0.65

vernment

-0.63

 resumed

-0.62

 diaper

-0.61

tips

-0.56

Dys

-0.56

BuyableInstoreAndOnline

-0.56

Phill

-0.55

iatus

-0.55

POSITIVE LOGITS

requires

0.97

must

0.87

apo

0.86

 requires

0.84

 must

0.81

 require

0.77

ru

0.74

needs

0.72

need

0.69

Firstly

0.68

Activations Density 0.758%

formatted text samples within the token

No Comments

No Known Activations