INDEX

Explanations

references to national events or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

nice

-0.19

thing

-0.16

-Methods

-0.15

sse

-0.15

rij

-0.14

ory

-0.14

inand

-0.14

 nice

-0.14

ocha

-0.14

á»«a

-0.13

POSITIVE LOGITS

/reg

0.21

/global

0.21

/local

0.21

LEGRO

0.18

/world

0.18

ities

0.18

ization

0.17

-level

0.16

/state

0.16

ity

0.16

Activations Density 0.059%

references to national events or entities

No Comments

No Known Activations