INDEX

Explanations

phrases related to recommendation or suggestion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

iverse

-0.15

hq

-0.15

yre

-0.15

opa

-0.14

à¸§à¸ĩ

-0.14

wiki

-0.14

urdy

-0.14

ÑĨÑı

-0.13

ãĥ¼ãĥ³

-0.13

idades

-0.13

POSITIVE LOGITS

 location

0.20

 timing

0.18

timing

0.18

Timing

0.17

 morphology

0.17

 structure

0.16

 Location

0.16

 syntax

0.16

 ideology

0.16

 appearance

0.16

Activations Density 0.002%

phrases related to recommendation or suggestion

No Comments

No Known Activations