INDEX

Explanations

questions and discussions about unresolved mysteries or investigations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

unta

-0.16

Duy

-0.15

iesel

-0.15

Æ°á»Ľc

-0.14

vd

-0.14

à¤¿à¤¥

-0.14

 carg

-0.14

xic

-0.14

/cpu

-0.14

PCS

-0.13

POSITIVE LOGITS

 Twin

0.36

 Lynch

0.30

 twin

0.23

 Agent

0.23

 Lodge

0.22

 agents

0.22

 Agents

0.21

Agent

0.21

agents

0.20

Agents

0.20

Activations Density 0.006%

questions and discussions about unresolved mysteries or investigations

No Comments

No Known Activations