INDEX

Explanations

arguments and statements related to moral contradictions and hypocrisy

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

Ð¿Ð¾Ñģ

-0.17

ardless

-0.15

_TOO

-0.15

avir

-0.14

ullan

-0.14

_regeneration

-0.14

à¤¼

-0.14

imli

-0.14

iland

-0.14

ÄĽr

-0.14

POSITIVE LOGITS

 necessarily

0.20

 anymore

0.19

 itself

0.18

eting

0.17

Fle

0.16

ani

0.16

 actual

0.16

angu

0.15

ctor

0.15

nor

0.15

Activations Density 0.083%

arguments and statements related to moral contradictions and hypocrisy

No Comments

No Known Activations