INDEX

Explanations

exclamatory statements and claims related to communication and truthfulness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

.ManyToMany

-0.15

 startPosition

-0.14

_deinit

-0.14

McG

-0.14

 hero

-0.14

ìĨĮê°ľ

-0.14

ÑıÐ²Ð¸

-0.13

icast

-0.13

eme

-0.13

 Fatal

-0.13

POSITIVE LOGITS

 claims

0.18

 Claims

0.16

 claim

0.15

|_|

0.15

sen

0.14

ÐĴÐ¡

0.14

 Vine

0.14

IFO

0.14

utut

0.14

ãģ¾ãģ§

0.14

Activations Density 0.361%

exclamatory statements and claims related to communication and truthfulness

No Comments

No Known Activations