INDEX

Explanations

terms related to abundance or being full

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

ners

-0.16

erman

-0.16

-0.15

uk

-0.15

iri

-0.14

sto

-0.14

isy

-0.14

ifo

-0.14

aha

-0.14

ached

-0.14

POSITIVE LOGITS

 with

0.30

 withd

0.23

with

0.23

	with

0.21

 dengan

0.21

-full

0.21

 vá»Ľi

0.21

 withString

0.19

 avec

0.18

WithValue

0.18

Activations Density 0.032%

terms related to abundance or being full

No Comments

No Known Activations