INDEX

Explanations

requests for user interaction, specifically related to contacting a service or providing feedback

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ÑģÑĤÑĢ

-0.18

uzz

-0.18

-0.15

 governmental

-0.15

ed

-0.14

áº¥n

-0.14

 Frog

-0.14

scription

-0.14

POSITIVE LOGITS

iat

0.17

IVA

0.15

Atlantic

0.15

.vaadin

0.15

annes

0.15

inery

0.14

ãĤĩ

0.14

 Ulus

0.14

Î¸Î¹

0.14

 WCHAR

0.14

Activations Density 0.047%

requests for user interaction, specifically related to contacting a service or providing feedback

No Comments

No Known Activations