INDEX

Explanations

instances of numerical or code references in the text

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

ÏģÏĮ

-0.17

 Merk

-0.16

ãĥ¼ãĤ¿

-0.16

ADDE

-0.15

enu

-0.15

mez

-0.15

ijken

-0.14

 METH

-0.14

asz

-0.14

ru

-0.14

POSITIVE LOGITS

itt

0.14

apons

0.14

eeper

0.14

Ð¼ÐµÑĢ

0.14

ev

0.14

 Ventures

0.14

è¡Ĺ

0.13

_expire

0.13

Activations Density 0.028%

instances of numerical or code references in the text

No Comments

No Known Activations