INDEX

Explanations

terms related to drug classification and legislation, particularly regarding marijuana and other controlled substances

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

isman

-0.64

etimes

-0.62

amina

-0.61

athered

-0.57

lore

-0.57

awi

-0.56

orians

-0.56

mosp

-0.56

ayed

-0.55

aying

-0.55

POSITIVE LOGITS

 onwards

1.11

 onward

0.95

+,

0.89

bis

0.85

+.

0.84

âĢ³

0.77

III

0.76

XXX

0.76

XX

0.75

Activations Density 1.450%

terms related to drug classification and legislation, particularly regarding marijuana and other controlled substances

No Comments

No Known Activations