INDEX

Explanations

phrases related to providing information or explanations

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

aired

-0.69

 Instr

-0.67

idal

-0.66

ä½ľ

-0.65

luaj

-0.63

Joined

-0.62

 throats

-0.60

ãĤ¡

-0.59

goo

-0.58

ocused

-0.58

POSITIVE LOGITS

 lest

0.83

 secondly

0.81

ÂłÂłÂł

0.79

how

0.75

 that

0.74

 caveats

0.73

:]

0.69

enance

0.68

 incidentally

0.68

WHY

0.67

Activations Density 0.080%

phrases related to providing information or explanations

No Comments

No Known Activations