INDEX

Explanations

punctuations or commas used 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 @ 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

and

-0.19

the

-0.18

but

-0.18

irt

-0.17

↵

-0.17

 Scha

-0.15

ep

-0.15

ess

-0.15

...↵

-0.15

âĢ¦

-0.14

POSITIVE LOGITS

utow

0.16

edBy

0.16

reesome

0.15

ilogy

0.15

ObjectName

0.15

GetMethod

0.15

usan

0.14

 ,↵↵

0.14

-,

0.14

NodeType

0.14

Activations Density 0.409%

punctuations or commas used in various contexts

No Comments

No Known Activations