INDEX

Explanations

entities related to historical and cultural references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

 ãĢĪ

-0.16

```

-0.16

 âĺĨ

-0.15

ĥģ

-0.15

æŁ±

-0.14

 #####

-0.14

ental

-0.14

Dit

-0.13

MOOTH

-0.13

xea

-0.13

POSITIVE LOGITS

ï»¿↵

0.17

 -↵

0.16

**

0.16

longleftrightarrow

0.16

	TokenName

0.16

‘s

0.15

ċ

0.15

ï»¿#

0.15

0.14

Activations Density 0.047%

entities related to historical and cultural references

No Comments

No Known Activations