INDEX

Explanations

references to items or actions that are "above" a certain level or threshold

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

luck

-0.16

 Rowe

-0.15

abant

-0.15

uers

-0.15

 Slee

-0.15

 Bridges

-0.14

net

-0.14

dle

-0.14

SAC

-0.14

 sack

-0.13

POSITIVE LOGITS

LastError

0.17

ScreenState

0.16

TestClass

0.15

 Cher

0.15

enson

0.15

reek

0.15

ukkan

0.14

Ã©gor

0.14

.Toolkit

0.14

/down

0.14

Activations Density 0.021%

references to items or actions that are "above" a certain level or threshold

No Comments

No Known Activations