INDEX

Explanations

instances of direct speech and conversation fragments

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

interstitial

-0.84

grave

-0.82

 firsthand

-0.68

bra

-0.67

 Retrieved

-0.65

listed

-0.65

HAM

-0.62

LEY

-0.61

ricanes

-0.61

enge

-0.61

POSITIVE LOGITS

 merely

1.07

 simply

0.97

just

0.91

 concentrate

0.88

 just

0.88

 suffice

0.88

Instead

0.80

pure

0.78

aceae

0.76

 Just

0.75

Activations Density 0.566%

instances of direct speech and conversation fragments

No Comments

No Known Activations