INDEX

Explanations

phrases that express continuity or repetitive actions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

 Emirates

-0.67

 Levant

-0.66

END

-0.66

 Retail

-0.64

çīĪ

-0.63

FK

-0.63

 Else

-0.63

å£«

-0.63

Adds

-0.61

qual

-0.60

POSITIVE LOGITS

 relax

0.89

 passively

0.87

 silently

0.84

 stared

0.80

 gazing

0.79

 surrounded

0.78

ertodd

0.77

 gazed

0.77

 listened

0.76

 stare

0.75

Activations Density 0.109%

phrases that express continuity or repetitive actions

No Comments

No Known Activations

phrases that express continuity or repetitive actions

No Comments

No Known Activations