INDEX

Explanations

the occurrence of the word "tw" and its variations, indicating a focus on social media references, particularly related to Twitter

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

vetica

-0.16

ung

-0.15

ØªØ§Ø¨

-0.15

jav

-0.15

Ð¾Ð²Ð°ÑĢ

-0.14

ÙĦÙĬÙĩ

-0.14

UNG

-0.14

hyp

-0.14

ÑĤÐµÑĢÐ¸

-0.14

istrovstvÃŃ

-0.14

POSITIVE LOGITS

viso

0.20

åĽ´

0.17

Ã³r

0.15

ided

0.15

assi

0.15

nee

0.14

ìłĢ

0.14

etik

0.14

0.13

esor

0.13

Activations Density 0.012%

the occurrence of the word "tw" and its variations, indicating a focus on social media references, particularly related to Twitter

No Comments

No Known Activations