INDEX

Explanations

references to specific years and their significance

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

 guiActiveUn

-0.67

atem

-0.62

emort

-0.60

ataka

-0.60

haar

-0.59

gently

-0.59

Fac

-0.58

ufficient

-0.58

lev

-0.58

 Rules

-0.57

POSITIVE LOGITS

nings

0.95

 outing

0.80

 opener

0.77

————————

0.75

long

0.73

(>

0.71

 nonetheless

0.71

coat

0.70

../

0.70

————

0.70

Activations Density 0.067%

references to specific years and their significance

No Comments

No Known Activations