INDEX

Explanations

instances of minimization or downplaying of situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

aise

-0.18

dorf

-0.17

idd

-0.15

obar

-0.15

hd

-0.15

oci

-0.15

ica

-0.14

kla

-0.14

POSITIVE LOGITS

ventus

0.16

DTD

0.15

Ã¡v

0.15

usta

0.15

tons

0.15

UNT

0.14

ediator

0.14

/Dk

0.14

 strchr

0.14

ekk

0.14

Activations Density 0.158%

instances of minimization or downplaying of situations

No Comments

No Known Activations