INDEX

Explanations

terms or phrases in a language that is likely not in English, possibly focusing on specific cultural or regional references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

anian

-0.16

ÑıÑĩ

-0.16

ÐĽÐ¬

-0.15

åĺĽ

-0.15

nnen

-0.15

isci

-0.14

getStatusCode

-0.14

ÑģÑĤÑĭ

-0.14

Ãºde

-0.14

Nel

-0.14

POSITIVE LOGITS

 Ð¸Ð¼Ñĥ

0.17

 ÐµÐ»ÐµÐº

0.16

 ÐµÐ»ÐµÐºÑĤÑĢÐ¾Ð½

0.16

ÑĬ

0.15

gate

0.15

ato

0.15

.Engine

0.15

 Ð½ÐµÐ´ÐµÐ»

0.15

bars

0.15

ÐµÑī

0.15

Activations Density 0.008%

terms or phrases in a language that is likely not in English, possibly focusing on specific cultural or regional references

No Comments

No Known Activations