INDEX

Explanations

instances of the word "reference" and its variations, indicating a focus on citations or references in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

iron

-0.16

ForeignKey

-0.15

 frank

-0.15

ern

-0.15

ards

-0.15

arr

-0.15

arily

-0.14

eenth

-0.14

 chá»ĭu

-0.14

Ð¸Ð²ÑĪÐ¸

-0.14

POSITIVE LOGITS

resher

0.18

izes

0.18

luÅ¾

0.16

coni

0.15

oldem

0.15

exual

0.15

ential

0.15

ÎµÎ¹Î±

0.15

attles

0.14

utable

0.14

Activations Density 0.058%

instances of the word "reference" and its variations, indicating a focus on citations or references in the text

No Comments

No Known Activations