INDEX

Explanations

words indicating frequency or repetition

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

anmar

-0.70

agra

-0.68

sterdam

-0.67

 citiz

-0.66

heid

-0.65

uckland

-0.65

 Integrity

-0.64

gebra

-0.64

ENGTH

-0.64

owell

-0.63

POSITIVE LOGITS

entimes

1.29

 than

1.26

than

1.03

 occurrence

0.82

 times

0.79

 occurrences

0.77

times

0.77

likely

0.75

theless

0.75

 Than

0.71

Activations Density 0.016%

words indicating frequency or repetition

No Comments

No Known Activations