INDEX

Explanations

mentions of "dollars" or "cents"

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

hess

-0.72

RET

-0.64

 Journals

-0.63

baugh

-0.62

forth

-0.61

tal

-0.61

////////////////

-0.61

 Bett

-0.61

 giveaways

-0.60

acebook

-0.60

POSITIVE LOGITS

 month

1.11

 week

1.04

 year

0.99

 capita

0.94

 fortnight

0.92

month

0.91

 hour

0.91

day

0.90

 weekday

0.88

 pound

0.88

Activations Density 0.486%

mentions of "dollars" or "cents"

No Comments

No Known Activations