INDEX

Explanations

instances of the word "turn" or its various forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

emas

-0.16

umd

-0.15

HashCode

-0.15

oop

-0.15

bed

-0.15

imum

-0.15

lant

-0.15

hoot

-0.15

rikes

-0.15

langs

-0.15

POSITIVE LOGITS

pike

0.27

tables

0.25

stile

0.24

coat

0.23

-around

0.22

around

0.22

 heads

0.21

 away

0.21

ips

0.20

 loose

0.19

Activations Density 0.027%

instances of the word "turn" or its various forms

No Comments

No Known Activations

instances of the word "turn" or its various forms

No Comments

No Known Activations