INDEX

Explanations

instances of the word "control" in relation to power dynamics

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

ibal

-0.16

éĽ

-0.14

ussen

-0.14

ornings

-0.14

allen

-0.14

EDIATE

-0.14

æĮĻ

-0.14

oris

-0.14

arium

-0.14

åĪ·

-0.13

POSITIVE LOGITS

iser

0.16

 Platt

0.15

ufe

0.14

å½¦

0.14

/browse

0.14

ervo

0.14

lun

0.14

nde

0.13

imb

0.13

Activations Density 0.034%

instances of the word "control" in relation to power dynamics

No Comments

No Known Activations