INDEX

Explanations

references to historical documents and related figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ì²Ń

-0.16

ombo

-0.15

anche

-0.15

 Sham

-0.15

/downloads

-0.14

downloads

-0.13

uilder

-0.13

ANEL

-0.13

 Cyrus

-0.13

agem

-0.13

POSITIVE LOGITS

ube

0.15

ikel

0.14

 personel

0.14

vala

0.14

è§

0.14

 Casting

0.14

ehr

0.13

omi

0.13

kv

0.13

oor

0.13

Activations Density 0.011%

references to historical documents and related figures

No Comments

No Known Activations