INDEX

Explanations

phrases related to definitions and coding functions

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

 Claus

-0.16

acers

-0.15

ãģ¨ãģĵãĤį

-0.15

ubits

-0.14

 nowhere

-0.14

burgh

-0.14

.lu

-0.14

rieb

-0.13

cola

-0.13

cedure

-0.13

POSITIVE LOGITS

TouchUpInside

0.17

 SPDX

0.15

ikit

0.15

 Burst

0.14

 íķ´ê²°

0.14

_hint

0.14

unde

0.14

enda

0.14

anker

0.14

Activations Density 0.052%

phrases related to definitions and coding functions

No Comments

No Known Activations