INDEX

Explanations

elements related to conflict and action

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

irie

-0.18

hind

-0.16

avou

-0.16

 Hind

-0.16

.nano

-0.15

zew

-0.15

VERR

-0.15

ophon

-0.14

à¤¹à¤¨

-0.14

POSITIVE LOGITS

olean

0.15

lette

0.15

cx

0.15

äºķ

0.15

 withhold

0.14

igit

0.14

lexical

0.14

à¸ĺà¸£à¸£à¸¡

0.14

umbled

0.14

ë²¨

0.14

Activations Density 0.459%

elements related to conflict and action

No Comments

No Known Activations