INDEX

Explanations

phrases indicating user engagement with content, particularly "read more" prompts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

/tos

-0.14

oro

-0.14

igid

-0.14

acea

-0.14

orns

-0.14

ches

-0.14

 packing

-0.14

olumn

-0.14

web

-0.13

orn

-0.13

POSITIVE LOGITS

 Gamb

0.16

fad

0.15

ormsg

0.14

ELLOW

0.14

.sheet

0.14

Å¡li

0.14

ÏįÏĢ

0.14

Zuk

0.13

ERV

0.13

mpar

0.13

Activations Density 0.028%

phrases indicating user engagement with content, particularly "read more" prompts

No Comments

No Known Activations