INDEX

Explanations

specific brand names and product names in the context of food and beverages

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

sembly

-0.81

actionDate

-0.77

PDATE

-0.75

 carbohyd

-0.74

inarily

-0.74

GoldMagikarp

-0.73

ccording

-0.72

depending

-0.71

etheless

-0.70

 fairness

-0.69

POSITIVE LOGITS

 Presents

1.02

 Orchestra

1.00

ensis

1.00

 Syndrome

0.95

Jr

0.91

aka

0.85

 Blues

0.84

 Company

0.83

III

0.82

II

0.82

Activations Density 0.122%

specific brand names and product names in the context of food and beverages

No Comments

No Known Activations