INDEX

Explanations

media outlets and news broadcasting channels

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĥ³ãĥĶ

-0.17

 dich

-0.17

Ear

-0.15

otron

-0.15

 Tort

-0.14

istributed

-0.14

 TORT

-0.14

andest

-0.14

_MO

-0.13

houette

-0.13

POSITIVE LOGITS

-TV

0.35

tv

0.20

 Channel

0.20

TV

0.20

 television

0.19

 radio

0.19

 channel

0.18

TV

0.18

Channel

0.17

_TV

0.17

Activations Density 0.023%

media outlets and news broadcasting channels

No Comments

No Known Activations