INDEX

Explanations

commands or prompts related to selection

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

aps

-0.16

 Alive

-0.15

onga

-0.15

 nues

-0.15

iliary

-0.14

idir

-0.14

 succeeded

-0.14

-selection

-0.14

rescia

-0.13

-defense

-0.13

POSITIVE LOGITS

ivity

0.27

ively

0.23

ive

0.21

IVE

0.20

ives

0.18

IVITY

0.18

ieve

0.18

iveness

0.17

'gc

0.16

iven

0.16

Activations Density 0.010%

commands or prompts related to selection

No Comments

No Known Activations