INDEX

Explanations

instances of dialogue or conversation, particularly focusing on responses and the act of replying in a communicative context

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

wald

-0.16

quiet

-0.15

emand

-0.15

.setScene

-0.14

obia

-0.14

repeat

-0.14

QUI

-0.14

 trib

-0.13

BOVE

-0.13

_REPEAT

-0.13

POSITIVE LOGITS

 replies

0.36

 reply

0.36

 response

0.30

 responses

0.30

 replied

0.29

 Replies

0.29

Reply

0.28

reply

0.27

 responded

0.27

response

0.26

Activations Density 0.073%

instances of dialogue or conversation, particularly focusing on responses and the act of replying in a communicative context

No Comments

No Known Activations