INDEX

Explanations

references to digital content and online interactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

 âĹĦ

-0.18

ocs

-0.16

]âĢı

-0.15

Ø§ØŃÛĮ

-0.14

Council

-0.14

.bias

-0.14

 Corrections

-0.13

orent

-0.13

emente

-0.13

hap

-0.13

POSITIVE LOGITS

addon

0.15

foy

0.15

Å¯r

0.15

isl

0.14

¼åĲĪ

0.14

 pair

0.14

ä¼¯

0.14

agner

0.14

enn

0.14

Activations Density 0.012%

references to digital content and online interactions

No Comments

No Known Activations