INDEX

Explanations

terms related to capabilities and possibilities

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

Și

-0.69

 unlaw

-0.61

 theolog

-0.52

Dimensi

-0.51

 egli

-0.51

 robus

-0.50

Să

-0.50

 emphat

-0.50

 exorbit

-0.50

 superintend

-0.49

POSITIVE LOGITS

able

0.79

ABLE

0.73

vable

0.73

atable

0.71

iable

0.69

gable

0.69

pable

0.66

ifiable

0.63

ishable

0.61

mable

0.61

Activations Density 0.200%

terms related to capabilities and possibilities

No Comments

No Known Activations