INDEX

Explanations

expressions of disbelief or critical evaluation of common narratives

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

unu

-0.15

ught

-0.15

esson

-0.15

UMP

-0.14

igan

-0.14

ollar

-0.14

 Earth

-0.14

 Economist

-0.14

gba

-0.13

os

-0.13

POSITIVE LOGITS

ë§Į

0.16

adox

0.15

ohn

0.15

afort

0.14

redient

0.14

mmc

0.14

TEX

0.14

.scalablytyped

0.14

olean

0.13

pond

0.13

Activations Density 0.032%

expressions of disbelief or critical evaluation of common narratives

No Comments

No Known Activations