INDEX

Explanations

interactions and relationships among individuals

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Girlfriend

-0.15

 ÐºÐ¾Ð¼Ð¿Ð»ÐµÐº

-0.14

 granddaughter

-0.14

lover

-0.14

aget

-0.14

 Glover

-0.14

avec

-0.14

enco

-0.14

ewolf

-0.14

FirstChild

-0.14

POSITIVE LOGITS

 guys

0.56

 boys

0.55

 fell

0.55

gu

0.50

 Guys

0.43

 fellows

0.41

boys

0.41

 Boys

0.41

Gu

0.39

Activations Density 0.201%

interactions and relationships among individuals

No Comments

No Known Activations