INDEX

Explanations

structural components of code or markup

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

erdem

-0.16

odel

-0.15

 Niet

-0.15

_Att

-0.15

 Depression

-0.14

fully

-0.14

nee

-0.14

paces

-0.14

oret

-0.14

lessly

-0.13

POSITIVE LOGITS

ADED

0.16

ch

0.15

ooth

0.14

hetto

0.14

 Nacht

0.14

iolet

0.14

olest

0.14

 strang

0.14

_episode

0.13

ule

0.13

Activations Density 0.037%

structural components of code or markup

No Comments

No Known Activations