INDEX

Explanations

phrases and words that imply a sense of obligation or responsibility

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

å¹²

-0.18

 Herc

-0.15

 å¹

-0.14

æĺĮ

-0.14

 Bowen

-0.14

 Chow

-0.14

StateException

-0.14

inox

-0.14

ÅĤe

-0.14

lant

-0.14

POSITIVE LOGITS

vo

0.16

lingen

0.16

oins

0.15

ive

0.15

ken

0.15

loor

0.15

.Builder

0.14

 Depths

0.14

arges

0.14

jon

0.14

Activations Density 0.005%

phrases and words that imply a sense of obligation or responsibility

No Comments

No Known Activations