INDEX

Explanations

conversations and interactions involving meetings and discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 lookout

-0.15

infeld

-0.15

.Accessible

-0.15

ereo

-0.14

acios

-0.14

opard

-0.14

kov

-0.14

 kepada

-0.13

iyan

-0.13

ÑĢÐµÑī

-0.13

POSITIVE LOGITS

 face

0.60

face

0.47

 Face

0.44

Face

0.40

-face

0.39

 FACE

0.33

.face

0.33

 faces

0.31

_face

0.30

éĿ¢

0.29

Activations Density 0.224%

conversations and interactions involving meetings and discussions

No Comments

No Known Activations