INDEX

Explanations

references to educational grades and age classifications

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

TRS

-0.17

ibble

-0.17

vent

-0.15

 Redistributions

-0.15

alt

-0.15

fusion

-0.15

eric

-0.15

onta

-0.14

 massaggi

-0.14

 Torch

-0.14

POSITIVE LOGITS

æŀľ

0.17

ONGL

0.15

-to

0.15

ouse

0.15

åĪ°

0.15

 aÅ¾

0.15

Ä±ÅŁÄ±k

0.15

bis

0.15

Activations Density 0.169%

references to educational grades and age classifications

No Comments

No Known Activations