INDEX

Explanations

references to windows in various contexts

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

inerary

-0.17

vÃ¤

-0.16

ÑĢÑĸÐ´

-0.16

cul

-0.15

lej

-0.15

Ãªu

-0.15

TimeString

-0.15

iedad

-0.14

leur

-0.14

ideographic

-0.14

POSITIVE LOGITS

pane

0.37

ill

0.28

ills

0.27

pan

0.24

-pane

0.24

æĪ·

0.24

ed

0.23

 sill

0.20

-shopping

0.20

ous

0.20

Activations Density 0.023%

references to windows in various contexts

No Comments

No Known Activations