INDEX

Explanations

numerical time-related expressions followed by the word "left"

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.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

Soc

-0.69

Jr

-0.62

 Orth

-0.61

pecially

-0.60

 Harmony

-0.58

ħĭ

-0.58

attribute

-0.58

 antit

-0.57

 Loud

-0.57

 Associ

-0.55

POSITIVE LOGITS

 left

1.32

 remaining

1.27

 leftover

1.03

 elapsed

0.91

 separating

0.90

 until

0.90

 allotted

0.88

 between

0.81

 till

0.81

iggle

0.78

Activations Density 0.092%

numerical time-related expressions followed by the word "left"

No Comments

No Known Activations