INDEX

Explanations

proper nouns, specifically names of individuals and places

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

adesh

-0.17

unes

-0.16

forman

-0.16

inct

-0.15

encing

-0.15

emand

-0.15

bette

-0.15

mani

-0.15

unist

-0.15

igel

-0.15

POSITIVE LOGITS

 Arms

0.15

eden

0.14

 nowhere

0.14

 Arthur

0.14

inv

0.14

REATED

0.14

åĿ¡

0.14

 family

0.13

hausen

0.13

Activations Density 0.103%

proper nouns, specifically names of individuals and places

No Comments

No Known Activations