INDEX

Explanations

terms and phrases related to calculations and numerical analysis

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

á»¡

-0.17

EntryPoint

-0.15

ogl

-0.15

ynth

-0.15

_regular

-0.15

 Caller

-0.15

edium

-0.14

 GenÃ§

-0.14

adata

-0.14

"><?=$

-0.14

POSITIVE LOGITS

 based

0.19

thon

0.16

 instead

0.16

based

0.16

andon

0.15

 tang

0.15

 basis

0.15

 criterion

0.15

-based

0.14

_based

0.14

Activations Density 0.155%

terms and phrases related to calculations and numerical analysis

No Comments

No Known Activations