INDEX

Explanations

references to navigation and direction in a narrative context

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

WN

-0.14

opro

-0.14

ioc

-0.14

 åī

-0.14

phia

-0.14

opo

-0.13

 sofort

-0.13

Ø¨Ø§ÙĨ

-0.13

cj

-0.13

liÄŁ

-0.13

POSITIVE LOGITS

 finally

0.18

finally

0.17

 halt

0.17

 reaches

0.16

inth

0.15

 llegar

0.15

ICLE

0.14

WL

0.14

 reached

0.14

rag

0.14

Activations Density 0.113%

references to navigation and direction in a narrative context

No Comments

No Known Activations