INDEX

Explanations

instances of the word "more" and its variations, indicating a focus on growth or increase

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

.scalablytyped

-0.18

ãĥ«ãĥķ

-0.17

ãĤ¤ãĤº

-0.15

ugins

-0.14

ÐµÐ³Ð¾ÑĢ

-0.14

riors

-0.14

ersed

-0.14

Versions

-0.14

 Retro

-0.14

Wunused

-0.14

POSITIVE LOGITS

 strat

0.15

æľĹ

0.14

ado

0.14

ivery

0.14

 assert

0.14

coop

0.14

ÐµÐ±

0.13

 Rodrigo

0.13

arena

0.13

.pub

0.13

Activations Density 0.231%

instances of the word "more" and its variations, indicating a focus on growth or increase

No Comments

No Known Activations