INDEX

Explanations

references to mathematical notation and variables

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

éģĩ

-0.14

:↵↵↵↵↵↵

-0.13

æ°ĹãģĮ

-0.13

coon

-0.13

ifer

-0.13

iro

-0.13

prend

-0.13

eden

-0.13

:^

-0.13

POSITIVE LOGITS

axter

0.15

inder

0.14

ourt

0.14

inel

0.14

ighter

0.13

reative

0.13

ensch

0.13

 angle

0.13

side

0.13

¬ģ

0.13

Activations Density 0.061%

references to mathematical notation and variables

No Comments

No Known Activations