INDEX

Explanations

references to programming constructs and methods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

oogle

-0.15

 Rocky

-0.14

 prepaid

-0.14

 Rolex

-0.14

lew

-0.14

vale

-0.14

 variant

-0.13

 crib

-0.13

_regularizer

-0.13

aria

-0.13

POSITIVE LOGITS

 Parse

0.39

Parse

0.37

 parse

0.30

(Parse

0.30

.Parse

0.28

parse

0.28

-parse

0.27

PARSE

0.27

_Parse

0.26

 Parses

0.26

Activations Density 0.001%

references to programming constructs and methods

No Comments

No Known Activations