INDEX

Explanations

references to educational or training tools and resources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ober

-0.15

adil

-0.14

ldkf

-0.14

/stdc

-0.14

ariance

-0.14

clave

-0.14

MOOTH

-0.14

andest

-0.14

ransition

-0.13

(disposing

-0.13

POSITIVE LOGITS

 inability

0.19

 ability

0.19

 lack

0.18

use

0.17

 absence

0.17

 being

0.16

 increased

0.15

 involvement

0.15

 having

0.14

etc

0.14

Activations Density 0.116%

references to educational or training tools and resources

No Comments

No Known Activations