INDEX

Explanations

references to frequency or instances of actions and conditions in various contexts

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

izo

-0.15

-original

-0.15

 Trou

-0.15

 Tran

-0.15

tons

-0.14

 Original

-0.14

 original

-0.14

azar

-0.14

original

-0.14

hood

-0.14

POSITIVE LOGITS

 once

1.02

once

0.91

 Once

0.81

Once

0.79

_once

0.64

 einmal

0.60

ä¸Ģæ¬¡

0.50

.once

0.50

 eens

0.47

 íķľë²Ī

0.46

Activations Density 0.072%

references to frequency or instances of actions and conditions in various contexts

No Comments

No Known Activations