INDEX

Explanations

phrases related to requirements or necessary actions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

gdala

-0.69

Zip

-0.67

 Guilty

-0.63

 Democr

-0.63

 Rost

-0.61

 Mamm

-0.61

izon

-0.60

SHIP

-0.58

cow

-0.56

 Varg

-0.55

POSITIVE LOGITS

lessly

1.11

 attention

0.85

 scrutiny

0.79

FINE

0.74

 repairs

0.72

0.71

 tweaking

0.70

 precedence

0.70

ENTION

0.69

 urgently

0.69

Activations Density 0.047%

phrases related to requirements or necessary actions

No Comments

No Known Activations