INDEX

Explanations

various forms of the prefix "re" indicating repetition or return

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

rey

-0.18

re

-0.17

reb

-0.16

rec

-0.15

res

-0.15

scar

-0.15

ru

-0.15

led

-0.15

aker

-0.14

oy

-0.14

POSITIVE LOGITS

 spons

0.22

becca

0.21

iew

0.20

ardon

0.20

chts

0.20

 cord

0.19

entrant

0.18

naissance

0.18

Leaf

0.17

ality

0.17

Activations Density 0.091%

various forms of the prefix "re" indicating repetition or return

No Comments

No Known Activations