INDEX

Explanations

specific words or terms related to emotional or psychological states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

och

-0.22

oj

-0.20

oq

-0.20

oS

-0.20

ov

-0.20

oad

-0.19

oc

-0.19

oz

-0.19

op

-0.19

POSITIVE LOGITS

ÐµÑģÑĤÐ²

0.35

ÐµÑģÑĤÐ²Ð¾

0.34

ÐµÑģÑĤÐ²Ð°

0.28

ÐµÐ³Ð¾

0.28

ÐµÐ¼Ñĥ

0.26

ÐµÐ¼

0.26

ÐµÐ²Ð¾

0.23

ÐµÐ¹

0.23

ÐµÐº

0.23

ÐµÑģ

0.22

Activations Density 0.023%

specific words or terms related to emotional or psychological states

No Comments

No Known Activations