INDEX

Explanations

terms related to critique or criticism

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ha

-0.19

idity

-0.17

ience

-0.15

HA

-0.15

erule

-0.15

ths

-0.15

iyah

-0.15

lds

-0.14

esco

-0.14

hurst

-0.14

POSITIVE LOGITS

icism

0.29

ters

0.29

ically

0.26

ter

0.24

Ã©ri

0.21

ics

0.21

iques

0.18

icial

0.18

izens

0.18

elpers

0.18

Activations Density 0.007%

terms related to critique or criticism

No Comments

No Known Activations