INDEX

Explanations

specific encoded symbols or characters resembling a foreign alphabet or script

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

onica

-0.16

-addon

-0.15

imos

-0.14

urai

-0.14

lets

-0.14

 Village

-0.14

Ñĳ

-0.14

riers

-0.14

íĨłíĨł

-0.14

¹

-0.13

POSITIVE LOGITS

 MENU

0.15

 nues

0.15

alf

0.15

BOOLE

0.15

pra

0.15

enis

0.14

opr

0.14

ysz

0.14

ESC

0.14

else

0.14

Activations Density 0.003%

specific encoded symbols or characters resembling a foreign alphabet or script

No Comments

No Known Activations