INDEX

Explanations

phrases expressing positive sentiments or affirmations

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

iasi

-0.18

-0.17

orgia

-0.16

ãĥ³ãĥ

-0.16

-0.15

æĳ¸

-0.15

å±¥

-0.14

Î¯Î¿Ïħ

-0.14

 Chore

-0.14

POSITIVE LOGITS

lien

0.14

spr

0.14

Äįer

0.14

 Casc

0.14

ough

0.14

 cascade

0.14

yen

0.14

sat

0.13

ole

0.13

sy

0.13

Activations Density 0.025%

phrases expressing positive sentiments or affirmations

No Comments

No Known Activations