INDEX

Explanations

instances of the word "Following," suggesting a focus on events or actions that occur in a sequence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

ses

-0.22

ÏĢÎ»Î±

-0.15

GetX

-0.15

sel

-0.14

 Tracker

-0.13

ÙĦÙĪ

-0.13

anness

-0.13

zim

-0.13

 afect

-0.13

_timing

-0.13

POSITIVE LOGITS

 receipt

0.20

ly

0.17

neath

0.16

raphics

0.16

 passage

0.15

 upon

0.15

æ¢°

0.15

alamat

0.15

LY

0.15

 completion

0.15

Activations Density 0.022%

instances of the word "Following," suggesting a focus on events or actions that occur in a sequence

No Comments

No Known Activations