INDEX

Explanations

words related to winning or achievements

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

ang

-0.15

paque

-0.15

lak

-0.14

wart

-0.14

 travel

-0.14

ensitive

-0.14

aque

-0.14

tip

-0.14

erson

-0.14

 details

-0.14

POSITIVE LOGITS

rite

0.16

aren

0.15

ry

0.14

lashes

0.14

flate

0.14

obox

0.14

ī´

0.14

Ð¼Ðµ

0.14

oyo

0.14

 ushort

0.14

Activations Density 0.020%

words related to winning or achievements

No Comments

No Known Activations