INDEX

Explanations

expressions related to self-improvement and motivation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

urd

-0.15

aki

-0.15

dispatcher

-0.15

iq

-0.14

HeaderValue

-0.14

oser

-0.14

pose

-0.14

 dispatcher

-0.14

RenderTarget

-0.14

ocal

-0.13

POSITIVE LOGITS

 wall

0.29

 dwelling

0.22

 succ

0.21

 internal

0.21

 Wall

0.21

wall

0.20

-wall

0.20

Wall

0.19

dw

0.19

 catast

0.19

Activations Density 0.387%

expressions related to self-improvement and motivation

No Comments

No Known Activations