INDEX

Explanations

concepts related to memory and forgetting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

.ali

-0.15

alom

-0.15

macros

-0.14

ATUS

-0.14

 Gang

-0.14

ocha

-0.14

coc

-0.13

.messaging

-0.13

ibur

-0.13

¬¬

-0.13

POSITIVE LOGITS

 memory

0.39

 Memory

0.34

memory

0.33

 retrieval

0.31

Memory

0.31

 recall

0.31

 memories

0.31

 MEMORY

0.30

-memory

0.30

 retention

0.30

Activations Density 0.033%

concepts related to memory and forgetting

No Comments

No Known Activations