INDEX

Explanations

references to URLs and links

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

bish

-0.16

plex

-0.15

olf

-0.15

amat

-0.14

Ìģ

-0.14

 Wing

-0.14

unc

-0.14

chef

-0.14

ritis

-0.13

\'

-0.13

POSITIVE LOGITS

 https

0.20

ISIBLE

0.17

https

0.16

www

0.15

ÙıØ³

0.15

_^

0.15

Prefab

0.15

eid

0.14

outu

0.14

_https

0.14

Activations Density 0.044%

references to URLs and links

No Comments

No Known Activations