INDEX

Explanations

statements related to events, plans, and occurrences

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

ãĥĨ

-0.76

ãĥĩ

-0.69

çĶ°

-0.68

æ°

-0.67

MY

-0.66

Offline

-0.63

und

-0.63

Border

-0.63

QB

-0.62

ãĤ©

-0.62

POSITIVE LOGITS

 nevertheless

1.84

 nonetheless

1.83

 suffice

1.25

 anecd

1.05

hey

1.00

 luckily

0.99

 fortunately

0.95

 certainly

0.93

 anyways

0.86

 still

0.84

Activations Density 1.821%

statements related to events, plans, and occurrences

No Comments

No Known Activations