INDEX

Explanations

keywords related to actions, contributions, and qualities within various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ONGO

-0.18

ÛĮÙĨÙĩ

-0.16

SED

-0.14

engo

-0.14

Segoe

-0.14

ongo

-0.14

olet

-0.14

atch

-0.14

ldre

-0.13

pel

-0.13

POSITIVE LOGITS

OCI

0.17

.alias

0.15

 absor

0.15

 thÃ¹

0.14

imated

0.14

ighton

0.14

abl

0.14

ji

0.14

Abr

0.14

ABCDEFGHIJKLMNOP

0.14

Activations Density 0.004%

keywords related to actions, contributions, and qualities within various contexts

No Comments

No Known Activations