INDEX

Explanations

references to the concept of "iki."

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

ÑĩÐµÐ½ÐºÐ¾

-0.16

agnar

-0.15

ektor

-0.15

à¤µà¤°

-0.15

reuse

-0.14

regon

-0.14

ictor

-0.14

paren

-0.14

ong

-0.14

æĻ¶

-0.13

POSITIVE LOGITS

oris

0.18

 Myers

0.16

.bb

0.15

 unpl

0.14

 Raleigh

0.14

ERGY

0.13

rium

0.13

anja

0.13

appa

0.13

Activations Density 0.006%

references to the concept of "iki."

No Known Activations