INDEX

Explanations

references to manual processes and data handling

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

ometr

-0.15

ismic

-0.15

lemetry

-0.15

sgi

-0.14

uple

-0.14

locker

-0.13

metro

-0.13

wargs

-0.13

 anonym

-0.13

_ALIGNMENT

-0.13

POSITIVE LOGITS

 manual

0.91

 manually

0.86

 Manual

0.75

manual

0.74

 MANUAL

0.70

Manual

0.69

_manual

0.57

 hand

0.55

/manual

0.52

.manual

0.50

Activations Density 0.247%

references to manual processes and data handling

No Comments

No Known Activations