INDEX

Explanations

phrases related to actions or events that have occurred

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

velt

-0.63

 Archdemon

-0.62

 endeavour

-0.58

iege

-0.58

 experiment

-0.57

 endeavor

-0.57

haus

-0.56

 annex

-0.56

 Britann

-0.55

 cardinal

-0.54

POSITIVE LOGITS

rid

1.44

tin

1.01

cloneembedreportprint

0.98

 acquainted

0.94

TING

0.90

 distracted

0.87

DragonMagazine

0.85

 bored

0.84

 sucked

0.84

aways

0.82

Activations Density 0.987%

phrases related to actions or events that have occurred

No Comments

No Known Activations