INDEX

Explanations

verbs related to inserting or injecting something

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_fs6144-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_6144

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

6,144

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

Ĥİ

-0.84

ģ«

-0.67

 ende

-0.61

 Machina

-0.60

fman

-0.59

 Correspond

-0.58

 tending

-0.58

Uncommon

-0.57

 Balanced

-0.56

 Princ

-0.55

POSITIVE LOGITS

 into

1.73

 INTO

1.61

 Into

1.51

into

1.48

 onto

1.24

tion

0.87

lish

0.81

lished

0.78

 molded

0.73

ibly

0.71

Activations Density 0.570%

verbs related to inserting or injecting something

No Comments

No Known Activations