INDEX

Explanations

indicators of programming language constructs and types

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 mainland

-0.16

ashboard

-0.15

 Stuart

-0.15

ãĥªãĥ³

-0.15

oha

-0.15

undy

-0.15

 retr

-0.15

ANNER

-0.14

ductor

-0.14

imoto

-0.14

POSITIVE LOGITS

vtk

0.33

vtk

0.31

 actor

0.20

.actor

0.19

 Actor

0.19

voi

0.18

=v

0.18

 actors

0.18

<v

0.17

 volume

0.17

Activations Density 0.013%

indicators of programming language constructs and types

No Comments

No Known Activations