INDEX

Explanations

references to memory and how experiences are remembered or forgotten

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

 Buckley

-0.15

Ã¡ch

-0.15

 Ø¯Ø³Øª

-0.14

 interpret

-0.14

mask

-0.14

ttp

-0.14

adores

-0.14

imat

-0.14

_mask

-0.13

theid

-0.13

POSITIVE LOGITS

 memory

0.36

 Memory

0.32

memory

0.31

 memor

0.31

 ê¸°ìĸµ

0.30

 remembered

0.30

Memory

0.29

 MEMORY

0.29

MEMORY

0.29

 Memor

0.28

Activations Density 0.154%

references to memory and how experiences are remembered or forgotten

No Comments

No Known Activations