INDEX

Explanations

references to changes or inconsistencies in performance outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

 Stuff

-0.15

invalid

-0.15

 Invalid

-0.15

asel

-0.15

à¸Ńà¸ĩà¸Īà¸²à¸ģ

-0.14

 incompetence

-0.14

 derec

-0.14

akter

-0.14

 clas

-0.14

 Heck

-0.14

POSITIVE LOGITS

 variable

0.42

 patch

0.36

 mixed

0.36

mixed

0.36

 Variable

0.35

patch

0.33

-variable

0.33

Mixed

0.32

variable

0.32

Variable

0.32

Activations Density 0.152%

references to changes or inconsistencies in performance outcomes

No Comments

No Known Activations