INDEX

Explanations

terms related to programming or technical components

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

antha

-0.15

amburger

-0.15

elix

-0.14

akest

-0.14

urat

-0.14

arness

-0.14

idon

-0.14

unu

-0.13

imax

-0.13

dae

-0.13

POSITIVE LOGITS

Bloc

0.16

 Ð´ÐµÐ»Ð°

0.15

 Watson

0.15

Wat

0.14

 Ð¾ÑĤÐ´ÐµÐ»

0.13

 Miranda

0.13

ãĥĥãĤ°

0.13

=$((

0.13

ê¹

0.13

ëĤĺëĿ¼

0.13

Activations Density 0.005%

terms related to programming or technical components

No Comments

No Known Activations