INDEX

Explanations

specific nouns, especially related to locations, concepts, and notable individuals

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

Ð½Ð¾ÑģÑĤ

-0.15

escal

-0.15

ãĥ¼ãĥ©

-0.15

assel

-0.15

ÑĥÐ¹

-0.15

agens

-0.15

çĥĪ

-0.14

eroon

-0.14

 Ð²ÑģÐµÐ³Ð¾

-0.14

POSITIVE LOGITS

 Consort

0.14

ose

0.14

lamaz

0.14

idente

0.14

ibile

0.13

 Bernstein

0.13

 Revel

0.13

uer

0.13

ìĿ´ìĬ¤

0.13

worthy

0.13

Activations Density 0.071%

specific nouns, especially related to locations, concepts, and notable individuals

No Comments

No Known Activations