INDEX

Explanations

locations, especially names of places and regions

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

adio

-0.18

pun

-0.16

atte

-0.14

uling

-0.14

fruit

-0.14

lik

-0.14

ade

-0.14

TypeEnum

-0.14

azzi

-0.14

ful

-0.13

POSITIVE LOGITS

 deser

0.16

ainen

0.15

anness

0.15

CTest

0.15

marvin

0.14

ecz

0.14

reme

0.13

LEGRO

0.13

Ð¼Ñĭ

0.13

Activations Density 0.018%

locations, especially names of places and regions

No Comments

No Known Activations