INDEX

Explanations

technical instructions or explanations

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

urated

-0.80

ocused

-0.73

ravel

-0.72

aired

-0.70

body

-0.67

hung

-0.67

 integ

-0.67

oing

-0.65

luaj

-0.65

und

-0.63

POSITIVE LOGITS

 though

0.82

 there

0.71

WHY

0.69

adays

0.68

 however

0.68

 caveats

0.67

 incidentally

0.67

 lest

0.66

 THERE

0.65

Activations Density 2.495%

technical instructions or explanations

No Comments

No Known Activations