INDEX

Explanations

words related to possession and individuality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

fen

-0.15

iban

-0.15

apers

-0.14

urum

-0.14

Bah

-0.14

Mig

-0.14

antino

-0.13

Wor

-0.13

uras

-0.13

POSITIVE LOGITS

 accord

0.16

stÃ¤nd

0.16

azen

0.16

ledge

0.15

esty

0.15

ritch

0.15

-ÑĤÐ°ÐºÐ¸

0.14

 personel

0.14

-brand

0.14

/internal

0.14

Activations Density 0.049%

words related to possession and individuality

No Comments

No Known Activations