INDEX

Explanations

mathematical expressions and notations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ugas

-0.16

 Hammond

-0.14

hari

-0.14

errer

-0.14

rella

-0.13

/assert

-0.13

reu

-0.13

olini

-0.13

 Econom

-0.13

rellas

-0.13

POSITIVE LOGITS

right

0.47

 right

0.41

Right

0.32

RIGHT

0.32

-right

0.32

 Right

0.31

 RIGHT

0.31

_right

0.29

_Right

0.27

(right

0.26

Activations Density 0.081%

mathematical expressions and notations

No Comments

No Known Activations