INDEX

Explanations

instances of the phrase "to be" with varying intensities, indicating a focus on existential or descriptive statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

ibox

-0.20

 borne

-0.15

vang

-0.14

ì±ħ

-0.14

dete

-0.14

-un

-0.14

vinc

-0.13

Ńå·ŀ

-0.13

oglob

-0.13

 meter

-0.13

POSITIVE LOGITS

Cum

0.19

Cum

0.16

 Wenger

0.15

cum

0.15

Ã§on

0.15

meric

0.14

[rand

0.14

andest

0.14

asting

0.14

inear

0.14

Activations Density 0.114%

instances of the phrase "to be" with varying intensities, indicating a focus on existential or descriptive statements

No Comments

No Known Activations