INDEX

Explanations

phrases related to observations or investigations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Mariners

-0.70

 Priest

-0.60

 Engineers

-0.58

 priesthood

-0.57

 NAACP

-0.56

 Adin

-0.56

 Zeit

-0.55

Fey

-0.55

DOT

-0.55

 Ment

-0.54

POSITIVE LOGITS

 beforehand

0.91

 throughout

0.83

 during

0.81

 dearly

0.79

.:

0.78

 before

0.75

 regarding

0.72

.</

0.71

 whilst

0.70

.?

0.70

Activations Density 1.572%

phrases related to observations or investigations

No Comments

No Known Activations