INDEX

Explanations

references to self-discovery and personal fulfillment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

orman

-0.19

 happen

-0.16

ÃŃk

-0.15

uke

-0.15

 Influence

-0.15

/Base

-0.14

ument

-0.14

ough

-0.14

loses

-0.14

chos

-0.14

POSITIVE LOGITS

 suits

0.22

 matters

0.22

 works

0.20

exc

0.20

 suites

0.19

 moves

0.18

 lights

0.18

 Matters

0.18

 interests

0.17

works

0.17

Activations Density 0.088%

references to self-discovery and personal fulfillment

No Comments

No Known Activations