INDEX

Explanations

discussions about hypocrisy and the coherence of beliefs and actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

ÑĤÑĢÐ°

-0.14

 sport

-0.14

lenen

-0.14

SEG

-0.14

bah

-0.14

 showc

-0.14

RunWith

-0.14

urgeon

-0.13

 physic

-0.13

 Anadolu

-0.13

POSITIVE LOGITS

 Alic

0.16

 meta

0.14

andard

0.14

.Meta

0.14

riel

0.14

meta

0.14

*>(&

0.14

mmas

0.14

opia

0.13

 cÃ©lib

0.13

Activations Density 0.140%

discussions about hypocrisy and the coherence of beliefs and actions

No Comments

No Known Activations