INDEX

Explanations

references to prior statements or mentions within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

orm

-0.14

ãĥĩãĥ«

-0.14

uti

-0.14

lcm

-0.14

ãĥ³ãĥĩ

-0.14

ey

-0.14

pir

-0.13

 volcano

-0.13

iqu

-0.13

POSITIVE LOGITS

UnderTest

0.16

/Dk

0.15

Balancer

0.15

zdy

0.15

ÏĩÎµÎ´ÏĮÎ½

0.15

undy

0.15

cord

0.14

zik

0.14

_shell

0.14

.wik

0.14

Activations Density 0.044%

references to prior statements or mentions within the text

No Comments

No Known Activations