INDEX

Explanations

names of places and related concepts

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 sappi

-1.31

 dises

-1.14

 roberto

-1.13

 sergio

-1.13

 vogli

-1.11

 gius

-1.11

 ?...

-1.11

 desir

-1.10

 soggior

-1.09

 ridu

-1.09

POSITIVE LOGITS

but

1.19

but

0.96

But

0.90

But

0.88

 nhưng

0.85

 whereas

0.81

BUT

0.80

 pero

0.79

0.77

Activations Density 0.896%

names of places and related concepts

No Comments

No Known Activations