INDEX

Explanations

mentions of rankings or positions in a list

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

bach

-0.15

pires

-0.15

ipop

-0.15

Aly

-0.15

anus

-0.14

ilda

-0.14

yte

-0.14

beck

-0.14

ÎºÎ¹

-0.13

HI

-0.13

POSITIVE LOGITS

ten

0.18

0.17

åįģ

0.17

ech

0.17

ï¼ĳï¼Ĳ

0.15

ivity

0.15

Ten

0.15

Ten

0.15

Coder

0.15

_ten

0.14

Activations Density 0.017%

mentions of rankings or positions in a list

No Comments

No Known Activations

mentions of rankings or positions in a list

No Comments

No Known Activations