INDEX

Explanations

proper nouns, especially names and titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

oten

-0.18

onya

-0.18

Ð»ÑıÐ½

-0.15

 Ø§ÙĦØ¹Ø¸

-0.15

unde

-0.15

 cons

-0.14

uden

-0.14

indow

-0.14

outu

-0.14

POSITIVE LOGITS

Å¡el

0.15

 bast

0.15

gaard

0.15

à¹Ģà¸ķà¸Ńà¸£

0.14

bury

0.14

REW

0.14

>:</

0.14

 indexing

0.13

-cli

0.13

 Competitive

0.13

Activations Density 0.006%

proper nouns, especially names and titles

No Comments

No Known Activations