INDEX

Explanations

locative terms and phrases indicating condition or state

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ikki

-0.15

otherwise

-0.15

inside

-0.15

 otherwise

-0.15

verified

-0.14

 inside

-0.14

amb

-0.14

vala

-0.14

 OTHERWISE

-0.14

Inside

-0.14

POSITIVE LOGITS

 alone

0.35

 especially

0.35

alone

0.34

 especialmente

0.29

 particularly

0.28

 Alone

0.28

å°¤

0.27

especially

0.26

 Especially

0.25

 besonders

0.25

Activations Density 0.008%

locative terms and phrases indicating condition or state

No Comments

No Known Activations