INDEX

Explanations

punctuation marks that indicate closure or transitions in thought, often associated with contrasting ideas or new statements

oai_token-act-pair · gpt-4o-mini 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

sing

-0.67

Enlarge

-0.61

emaker

-0.60

clus

-0.57

pert

-0.56

cel

-0.56

 mascul

-0.56

Chat

-0.55

 coun

-0.55

ãĥĺ

-0.54

POSITIVE LOGITS

 Anyway

0.99

 nevertheless

0.77

 Regardless

0.77

 Nonetheless

0.75

theless

0.75

BUT

0.74

but

0.73

 Nevertheless

0.72

Anyway

0.72

BUT

0.72

Activations Density 0.176%

punctuation marks that indicate closure or transitions in thought, often associated with contrasting ideas or new statements

No Comments

No Known Activations