INDEX

Explanations

references to specific attributes or metrics in datasets

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

lu

-0.15

 loads

-0.15

stant

-0.15

argo

-0.14

qli

-0.14

Loads

-0.14

olson

-0.14

lu

-0.14

ois

-0.14

Pocket

-0.14

POSITIVE LOGITS

ENCIES

0.16

PLY

0.16

ught

0.16

ernel

0.15

 spoiler

0.14

icha

0.14

 Edmund

0.14

leurs

0.14

Activations Density 0.007%

references to specific attributes or metrics in datasets

No Comments

No Known Activations