INDEX

Explanations

terms related to extraction and derivation processes

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

-0.16

asco

-0.15

aise

-0.15

ais

-0.14

 Marco

-0.14

YP

-0.14

 Choices

-0.14

onym

-0.14

 tack

-0.14

 THEORY

-0.14

POSITIVE LOGITS

ivism

0.18

andi

0.17

ivist

0.16

 dara

0.16

edd

0.16

bine

0.15

 sust

0.15

ãĥ¬ãĥĥãĥĪ

0.15

ILON

0.15

illard

0.15

Activations Density 0.038%

terms related to extraction and derivation processes

No Comments

No Known Activations