INDEX

Explanations

phrases indicating collaborations or partnerships

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L8R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.8.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

ligt

-0.15

åº

-0.15

acob

-0.15

ÑģÑĤÐµÑĢ

-0.14

pac

-0.14

ikki

-0.14

cline

-0.14

ccion

-0.14

intage

-0.14

 meis

-0.13

POSITIVE LOGITS

bru

0.16

0.15

 Constructs

0.14

rung

0.14

 partner

0.14

-ed

0.14

bru

0.13

 qualified

0.13

Activations Density 0.065%

phrases indicating collaborations or partnerships

No Comments

No Known Activations