INDEX

Explanations

code snippets or programming structures

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

ombok

-0.18

aney

-0.16

resh

-0.16

ORE

-0.15

zap

-0.15

nap

-0.15

iral

-0.14

ì²ł

-0.14

ulp

-0.14

ire

-0.14

POSITIVE LOGITS

ancia

0.15

 wich

0.15

aben

0.14

and

0.14

odom

0.14

_expiry

0.14

0.13

 Beat

0.13

AFX

0.13

Activations Density 0.092%

code snippets or programming structures

No Comments

No Known Activations