INDEX

Explanations

code-related terms and concepts associated with algorithms and data structures

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ogs

-0.16

|#

-0.14

[]↵

-0.14

ãģ®ä¸Ń

-0.14

 Yates

-0.14

/_

-0.14

(_

-0.14

&_

-0.14

((_

-0.13

:@

-0.13

POSITIVE LOGITS

_.

0.55

_,

0.50

_[

0.45

_;

0.45

_);↵

0.45

_)

0.45

_;↵

0.43

_->

0.43

_:

0.43

_)↵

0.43

Activations Density 0.253%

code-related terms and concepts associated with algorithms and data structures

No Comments

No Known Activations