INDEX

Explanations

proper nouns related to geography and politics

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

tops

-0.77

sequence

-0.75

fri

-0.66

yright

-0.64

dain

-0.63

othe

-0.61

iple

-0.60

 landlords

-0.60

asso

-0.59

 Alvin

-0.59

POSITIVE LOGITS

XT

0.82

 Yard

0.74

0.72

 Pavilion

0.70

ophobic

0.69

 Sunrise

0.68

 Clause

0.67

ite

0.66

icz

0.66

 Liberation

0.65

Activations Density 0.125%

proper nouns related to geography and politics

No Comments

No Known Activations