INDEX

Explanations

phrases related to conflicts, injustices, and tragedies

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

 Mariners

-0.66

 faint

-0.60

Fey

-0.57

 Priest

-0.56

 gears

-0.56

 Adin

-0.54

 NAACP

-0.54

 Zeit

-0.53

 coma

-0.53

 DMCA

-0.53

POSITIVE LOGITS

 beforehand

0.97

 during

0.87

.:

0.85

 throughout

0.80

.</

0.79

 before

0.79

 regarding

0.78

.?

0.76

abouts

0.75

 BEFORE

0.75

Activations Density 2.605%

phrases related to conflicts, injustices, and tragedies

No Comments

No Known Activations