INDEX

Explanations

instances of the name or references to Israel

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 yere

-0.15

andan

-0.15

ummer

-0.15

lacak

-0.15

atica

-0.14

 centrif

-0.14

posit

-0.14

 vestib

-0.14

ething

-0.13

POSITIVE LOGITS

esh

0.25

eh

0.24

ose

0.21

itz

0.20

IVO

0.20

ech

0.19

HV

0.19

is

0.18

ift

0.18

arden

0.18

Activations Density 0.007%

instances of the name or references to Israel

No Comments

No Known Activations

instances of the name or references to Israel

No Comments

No Known Activations