INDEX

Explanations

information related to movies, TV series, and productions

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_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

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

TPPStreamerBot

-0.95

statement

-0.86

Recommend

-0.80

inventoryQuantity

-0.78

endment

-0.78

Newsletter

-0.77

velop

-0.77

emark

-0.76

CSS

-0.76

Draft

-0.76

POSITIVE LOGITS

 protagonist

1.38

 heroine

1.17

 kidnapped

1.14

 narrator

1.13

 supernatural

1.09

 vampires

1.09

 flashbacks

1.08

 discovers

1.05

 adventures

1.04

 Dracula

1.04

Activations Density 6.259%

information related to movies, TV series, and productions

No Comments

No Known Activations