INDEX

Explanations

references to current systems, conditions, or policies and their impacts or characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

 previously

-0.19

 recently

-0.18

new

-0.17

 initially

-0.17

 lately

-0.17

 Previously

-0.16

 former

-0.16

 Ð½Ð¾Ð²Ð¾Ð³Ð¾

-0.16

 further

-0.15

ching

-0.14

POSITIVE LOGITS

/current

0.33

/original

0.28

-generation

0.24

 ones

0.22

 iteration

0.21

-version

0.21

-day

0.21

 incarnation

0.19

iteration

0.18

/new

0.18

Activations Density 0.208%

references to current systems, conditions, or policies and their impacts or characteristics

No Comments

No Known Activations