INDEX

Explanations

references to railway systems and associated terminology

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

Ã³c

-0.16

PPER

-0.15

 Bias

-0.15

pes

-0.14

 compr

-0.14

holes

-0.14

ãĥ³ãĥĨãĤ£

-0.14

.Func

-0.14

/manual

-0.14

POSITIVE LOGITS

EMU

0.23

 rake

0.20

DRM

0.20

 suburban

0.19

 coaches

0.18

ÙĪØ¯ÛĮ

0.18

ÂŃn

0.17

 Halk

0.16

 Coaching

0.16

 Coach

0.16

Activations Density 0.016%

references to railway systems and associated terminology

No Comments

No Known Activations