INDEX

Explanations

references to information or actions related to organization, comparison, and situational responses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

 Uncategorized

-0.15

maal

-0.14

Î»Î¬

-0.14

nutÃŃ

-0.13

zos

-0.13

á»ī

-0.13

ogan

-0.13

 Ð¿ÑĢÐ¾Ñĩ

-0.13

ackages

-0.12

ÅĤe

-0.12

POSITIVE LOGITS

 another

1.60

another

1.41

 Another

1.30

Another

1.23

åı¦

1.02

 otra

0.96

åı¦ä¸Ģ

0.96

 otro

0.91

 outra

0.79

 outro

0.75

Activations Density 0.741%

references to information or actions related to organization, comparison, and situational responses

No Comments

No Known Activations