INDEX

Explanations

snippets of code 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 @ 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

chn

-0.15

cot

-0.14

pos

-0.14

depth

-0.14

 Parcel

-0.13

clist

-0.13

ap

-0.13

opak

-0.13

POSITIVE LOGITS

ecycle

0.17

erah

0.17

verage

0.15

 SORT

0.15

Ã£i

0.15

ubre

0.15

eturn

0.15

finalize

0.15

à¸§à¸¡

0.14

odyn

0.14

Activations Density 0.098%

snippets of code or programming structures

No Comments

No Known Activations