INDEX

Explanations

repetitive phrases referring to thought or contemplation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 3-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L3R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.3.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

oker

-0.15

 branch

-0.15

atron

-0.14

iry

-0.14

roup

-0.13

INET

-0.13

 Dough

-0.13

 Forever

-0.13

\Modules

-0.13

POSITIVE LOGITS

yb

0.16

eza

0.14

 Hospitality

0.13

asion

0.13

Hav

0.13

eck

0.13

 RoundedRectangle

0.13

sÄ±nda

0.13

wil

0.13

 Retirement

0.13

Activations Density 0.030%

repetitive phrases referring to thought or contemplation

No Comments

No Known Activations