INDEX

Explanations

words related to maintenance and updates in various contexts

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

LastError

-0.17

oyer

-0.17

errer

-0.16

Î¼Î¬

-0.16

eria

-0.15

.metamodel

-0.15

jedn

-0.15

aque

-0.15

utta

-0.14

lea

-0.14

POSITIVE LOGITS

phans

0.16

GRE

0.15

inine

0.13

ìĦł

0.13

ients

0.13

 else

0.13

Sad

0.13

æł¹æľ¬

0.13

ocos

0.13

placement

0.13

Activations Density 0.376%

words related to maintenance and updates in various contexts

No Comments

No Known Activations