INDEX

Explanations

words indicating potential future actions or behaviors

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

âĢİ

-0.16

 sÃłng

-0.15

pone

-0.14

zzo

-0.14

[]{

-0.14

ome

-0.14

å®Ĺ

-0.14

sole

-0.13

atonin

-0.13

OME

-0.13

POSITIVE LOGITS

 Chall

0.17

andan

0.16

ukes

0.16

çĴĥ

0.15

rei

0.15

pau

0.15

ileaks

0.15

inta

0.15

ovna

0.14

ibri

0.14

Activations Density 0.000%

words indicating potential future actions or behaviors

No Comments

No Known Activations